{
  "best_global_step": 39200,
  "best_metric": 0.1431099772453308,
  "best_model_checkpoint": "saves/lntuning/gemma-3-1b-it/train_mnli_1744902583/checkpoint-39200",
  "epoch": 1.810792457955137,
  "eval_steps": 200,
  "global_step": 40000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00022635188664297517,
      "grad_norm": 10.110382080078125,
      "learning_rate": 4.999999876629946e-05,
      "loss": 4.8433,
      "num_input_tokens_seen": 8928,
      "step": 5
    },
    {
      "epoch": 0.00045270377328595035,
      "grad_norm": 10.270570755004883,
      "learning_rate": 4.999999375439123e-05,
      "loss": 4.8292,
      "num_input_tokens_seen": 16800,
      "step": 10
    },
    {
      "epoch": 0.0006790556599289255,
      "grad_norm": 9.259658813476562,
      "learning_rate": 4.9999984887169785e-05,
      "loss": 4.7415,
      "num_input_tokens_seen": 25088,
      "step": 15
    },
    {
      "epoch": 0.0009054075465719007,
      "grad_norm": 10.154754638671875,
      "learning_rate": 4.9999972164636506e-05,
      "loss": 4.6295,
      "num_input_tokens_seen": 33664,
      "step": 20
    },
    {
      "epoch": 0.0011317594332148758,
      "grad_norm": 9.575749397277832,
      "learning_rate": 4.999995558679334e-05,
      "loss": 4.8826,
      "num_input_tokens_seen": 40864,
      "step": 25
    },
    {
      "epoch": 0.001358111319857851,
      "grad_norm": 9.538041114807129,
      "learning_rate": 4.999993515364287e-05,
      "loss": 4.6253,
      "num_input_tokens_seen": 49280,
      "step": 30
    },
    {
      "epoch": 0.0015844632065008261,
      "grad_norm": 8.629034042358398,
      "learning_rate": 4.999991086518822e-05,
      "loss": 3.6662,
      "num_input_tokens_seen": 57312,
      "step": 35
    },
    {
      "epoch": 0.0018108150931438014,
      "grad_norm": 9.660565376281738,
      "learning_rate": 4.999988272143315e-05,
      "loss": 3.6102,
      "num_input_tokens_seen": 65248,
      "step": 40
    },
    {
      "epoch": 0.0020371669797867764,
      "grad_norm": 8.856003761291504,
      "learning_rate": 4.999985072238199e-05,
      "loss": 3.6502,
      "num_input_tokens_seen": 72928,
      "step": 45
    },
    {
      "epoch": 0.0022635188664297517,
      "grad_norm": 9.729303359985352,
      "learning_rate": 4.999981486803969e-05,
      "loss": 3.1371,
      "num_input_tokens_seen": 81440,
      "step": 50
    },
    {
      "epoch": 0.002489870753072727,
      "grad_norm": 9.5584716796875,
      "learning_rate": 4.999977515841176e-05,
      "loss": 3.37,
      "num_input_tokens_seen": 89216,
      "step": 55
    },
    {
      "epoch": 0.002716222639715702,
      "grad_norm": 9.001233100891113,
      "learning_rate": 4.9999731593504344e-05,
      "loss": 2.9114,
      "num_input_tokens_seen": 96736,
      "step": 60
    },
    {
      "epoch": 0.002942574526358677,
      "grad_norm": 8.449509620666504,
      "learning_rate": 4.999968417332415e-05,
      "loss": 2.8324,
      "num_input_tokens_seen": 104096,
      "step": 65
    },
    {
      "epoch": 0.0031689264130016523,
      "grad_norm": 7.94390869140625,
      "learning_rate": 4.999963289787848e-05,
      "loss": 2.5813,
      "num_input_tokens_seen": 112000,
      "step": 70
    },
    {
      "epoch": 0.0033952782996446275,
      "grad_norm": 8.673018455505371,
      "learning_rate": 4.999957776717526e-05,
      "loss": 2.3849,
      "num_input_tokens_seen": 119840,
      "step": 75
    },
    {
      "epoch": 0.0036216301862876028,
      "grad_norm": 9.040518760681152,
      "learning_rate": 4.9999518781222984e-05,
      "loss": 2.5184,
      "num_input_tokens_seen": 127520,
      "step": 80
    },
    {
      "epoch": 0.003847982072930578,
      "grad_norm": 7.344552040100098,
      "learning_rate": 4.9999455940030746e-05,
      "loss": 2.3708,
      "num_input_tokens_seen": 136448,
      "step": 85
    },
    {
      "epoch": 0.004074333959573553,
      "grad_norm": 7.8615922927856445,
      "learning_rate": 4.999938924360824e-05,
      "loss": 2.3064,
      "num_input_tokens_seen": 144128,
      "step": 90
    },
    {
      "epoch": 0.0043006858462165285,
      "grad_norm": 6.60552978515625,
      "learning_rate": 4.999931869196575e-05,
      "loss": 2.0194,
      "num_input_tokens_seen": 152032,
      "step": 95
    },
    {
      "epoch": 0.004527037732859503,
      "grad_norm": 6.072394371032715,
      "learning_rate": 4.999924428511416e-05,
      "loss": 1.7913,
      "num_input_tokens_seen": 159968,
      "step": 100
    },
    {
      "epoch": 0.004753389619502478,
      "grad_norm": 4.355900287628174,
      "learning_rate": 4.999916602306494e-05,
      "loss": 1.7528,
      "num_input_tokens_seen": 167712,
      "step": 105
    },
    {
      "epoch": 0.004979741506145454,
      "grad_norm": 6.5230560302734375,
      "learning_rate": 4.999908390583016e-05,
      "loss": 1.5532,
      "num_input_tokens_seen": 175680,
      "step": 110
    },
    {
      "epoch": 0.005206093392788429,
      "grad_norm": 5.3933424949646,
      "learning_rate": 4.999899793342247e-05,
      "loss": 1.781,
      "num_input_tokens_seen": 183776,
      "step": 115
    },
    {
      "epoch": 0.005432445279431404,
      "grad_norm": 4.599891662597656,
      "learning_rate": 4.999890810585516e-05,
      "loss": 1.8066,
      "num_input_tokens_seen": 191712,
      "step": 120
    },
    {
      "epoch": 0.005658797166074379,
      "grad_norm": 4.171250343322754,
      "learning_rate": 4.999881442314206e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 199200,
      "step": 125
    },
    {
      "epoch": 0.005885149052717354,
      "grad_norm": 2.8550899028778076,
      "learning_rate": 4.9998716885297617e-05,
      "loss": 1.3954,
      "num_input_tokens_seen": 207072,
      "step": 130
    },
    {
      "epoch": 0.00611150093936033,
      "grad_norm": 3.8727071285247803,
      "learning_rate": 4.999861549233688e-05,
      "loss": 1.6166,
      "num_input_tokens_seen": 215168,
      "step": 135
    },
    {
      "epoch": 0.0063378528260033045,
      "grad_norm": 4.462510108947754,
      "learning_rate": 4.999851024427548e-05,
      "loss": 1.2309,
      "num_input_tokens_seen": 223008,
      "step": 140
    },
    {
      "epoch": 0.00656420471264628,
      "grad_norm": 6.484999656677246,
      "learning_rate": 4.999840114112965e-05,
      "loss": 0.9951,
      "num_input_tokens_seen": 232320,
      "step": 145
    },
    {
      "epoch": 0.006790556599289255,
      "grad_norm": 2.3141961097717285,
      "learning_rate": 4.999828818291621e-05,
      "loss": 1.485,
      "num_input_tokens_seen": 240448,
      "step": 150
    },
    {
      "epoch": 0.00701690848593223,
      "grad_norm": 3.5569703578948975,
      "learning_rate": 4.999817136965259e-05,
      "loss": 1.0268,
      "num_input_tokens_seen": 248256,
      "step": 155
    },
    {
      "epoch": 0.0072432603725752056,
      "grad_norm": 4.206778526306152,
      "learning_rate": 4.9998050701356794e-05,
      "loss": 1.4578,
      "num_input_tokens_seen": 256320,
      "step": 160
    },
    {
      "epoch": 0.00746961225921818,
      "grad_norm": 2.631798028945923,
      "learning_rate": 4.999792617804744e-05,
      "loss": 1.247,
      "num_input_tokens_seen": 264160,
      "step": 165
    },
    {
      "epoch": 0.007695964145861156,
      "grad_norm": 2.296090602874756,
      "learning_rate": 4.9997797799743724e-05,
      "loss": 1.0509,
      "num_input_tokens_seen": 272800,
      "step": 170
    },
    {
      "epoch": 0.007922316032504132,
      "grad_norm": 2.585432767868042,
      "learning_rate": 4.999766556646545e-05,
      "loss": 1.2782,
      "num_input_tokens_seen": 280544,
      "step": 175
    },
    {
      "epoch": 0.008148667919147106,
      "grad_norm": 3.7724790573120117,
      "learning_rate": 4.9997529478232996e-05,
      "loss": 1.2233,
      "num_input_tokens_seen": 288736,
      "step": 180
    },
    {
      "epoch": 0.008375019805790081,
      "grad_norm": 3.3107829093933105,
      "learning_rate": 4.9997389535067365e-05,
      "loss": 0.8948,
      "num_input_tokens_seen": 296672,
      "step": 185
    },
    {
      "epoch": 0.008601371692433057,
      "grad_norm": 2.9943723678588867,
      "learning_rate": 4.999724573699012e-05,
      "loss": 1.1369,
      "num_input_tokens_seen": 305120,
      "step": 190
    },
    {
      "epoch": 0.008827723579076031,
      "grad_norm": 2.828873634338379,
      "learning_rate": 4.9997098084023457e-05,
      "loss": 1.1304,
      "num_input_tokens_seen": 312640,
      "step": 195
    },
    {
      "epoch": 0.009054075465719007,
      "grad_norm": 2.7086503505706787,
      "learning_rate": 4.999694657619013e-05,
      "loss": 1.0008,
      "num_input_tokens_seen": 320544,
      "step": 200
    },
    {
      "epoch": 0.009054075465719007,
      "eval_loss": 1.0760680437088013,
      "eval_runtime": 404.867,
      "eval_samples_per_second": 96.997,
      "eval_steps_per_second": 24.25,
      "num_input_tokens_seen": 320544,
      "step": 200
    },
    {
      "epoch": 0.009280427352361982,
      "grad_norm": 2.617600679397583,
      "learning_rate": 4.999679121351352e-05,
      "loss": 1.1977,
      "num_input_tokens_seen": 329344,
      "step": 205
    },
    {
      "epoch": 0.009506779239004956,
      "grad_norm": 2.726375102996826,
      "learning_rate": 4.9996631996017565e-05,
      "loss": 1.07,
      "num_input_tokens_seen": 338528,
      "step": 210
    },
    {
      "epoch": 0.009733131125647932,
      "grad_norm": 2.330773115158081,
      "learning_rate": 4.9996468923726835e-05,
      "loss": 1.1381,
      "num_input_tokens_seen": 346592,
      "step": 215
    },
    {
      "epoch": 0.009959483012290908,
      "grad_norm": 2.924706220626831,
      "learning_rate": 4.999630199666647e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 355360,
      "step": 220
    },
    {
      "epoch": 0.010185834898933883,
      "grad_norm": 4.431434154510498,
      "learning_rate": 4.999613121486222e-05,
      "loss": 1.1102,
      "num_input_tokens_seen": 363200,
      "step": 225
    },
    {
      "epoch": 0.010412186785576857,
      "grad_norm": 3.2073612213134766,
      "learning_rate": 4.999595657834041e-05,
      "loss": 0.9793,
      "num_input_tokens_seen": 370624,
      "step": 230
    },
    {
      "epoch": 0.010638538672219833,
      "grad_norm": 2.6591808795928955,
      "learning_rate": 4.999577808712798e-05,
      "loss": 1.111,
      "num_input_tokens_seen": 378368,
      "step": 235
    },
    {
      "epoch": 0.010864890558862809,
      "grad_norm": 5.437201976776123,
      "learning_rate": 4.999559574125244e-05,
      "loss": 1.1683,
      "num_input_tokens_seen": 386176,
      "step": 240
    },
    {
      "epoch": 0.011091242445505783,
      "grad_norm": 2.3993895053863525,
      "learning_rate": 4.9995409540741934e-05,
      "loss": 0.9573,
      "num_input_tokens_seen": 394496,
      "step": 245
    },
    {
      "epoch": 0.011317594332148758,
      "grad_norm": 3.0795271396636963,
      "learning_rate": 4.999521948562516e-05,
      "loss": 0.833,
      "num_input_tokens_seen": 402528,
      "step": 250
    },
    {
      "epoch": 0.011543946218791734,
      "grad_norm": 2.53849196434021,
      "learning_rate": 4.999502557593143e-05,
      "loss": 0.926,
      "num_input_tokens_seen": 410240,
      "step": 255
    },
    {
      "epoch": 0.011770298105434708,
      "grad_norm": 3.515901803970337,
      "learning_rate": 4.999482781169066e-05,
      "loss": 0.7532,
      "num_input_tokens_seen": 418240,
      "step": 260
    },
    {
      "epoch": 0.011996649992077684,
      "grad_norm": 2.4732863903045654,
      "learning_rate": 4.9994626192933324e-05,
      "loss": 0.8726,
      "num_input_tokens_seen": 426144,
      "step": 265
    },
    {
      "epoch": 0.01222300187872066,
      "grad_norm": 2.657785654067993,
      "learning_rate": 4.999442071969054e-05,
      "loss": 0.84,
      "num_input_tokens_seen": 434048,
      "step": 270
    },
    {
      "epoch": 0.012449353765363635,
      "grad_norm": 1.887690782546997,
      "learning_rate": 4.999421139199397e-05,
      "loss": 0.7608,
      "num_input_tokens_seen": 441728,
      "step": 275
    },
    {
      "epoch": 0.012675705652006609,
      "grad_norm": 1.9803428649902344,
      "learning_rate": 4.999399820987592e-05,
      "loss": 0.983,
      "num_input_tokens_seen": 449568,
      "step": 280
    },
    {
      "epoch": 0.012902057538649585,
      "grad_norm": 2.378300666809082,
      "learning_rate": 4.999378117336924e-05,
      "loss": 0.8555,
      "num_input_tokens_seen": 457408,
      "step": 285
    },
    {
      "epoch": 0.01312840942529256,
      "grad_norm": 2.354107141494751,
      "learning_rate": 4.9993560282507415e-05,
      "loss": 0.9838,
      "num_input_tokens_seen": 465376,
      "step": 290
    },
    {
      "epoch": 0.013354761311935534,
      "grad_norm": 1.3761131763458252,
      "learning_rate": 4.9993335537324495e-05,
      "loss": 0.9388,
      "num_input_tokens_seen": 473408,
      "step": 295
    },
    {
      "epoch": 0.01358111319857851,
      "grad_norm": 2.103109121322632,
      "learning_rate": 4.999310693785516e-05,
      "loss": 0.7288,
      "num_input_tokens_seen": 481760,
      "step": 300
    },
    {
      "epoch": 0.013807465085221486,
      "grad_norm": 2.85009765625,
      "learning_rate": 4.9992874484134653e-05,
      "loss": 1.1089,
      "num_input_tokens_seen": 489248,
      "step": 305
    },
    {
      "epoch": 0.01403381697186446,
      "grad_norm": 2.521477460861206,
      "learning_rate": 4.999263817619882e-05,
      "loss": 0.7391,
      "num_input_tokens_seen": 497344,
      "step": 310
    },
    {
      "epoch": 0.014260168858507435,
      "grad_norm": 3.015455961227417,
      "learning_rate": 4.9992398014084105e-05,
      "loss": 0.6872,
      "num_input_tokens_seen": 504832,
      "step": 315
    },
    {
      "epoch": 0.014486520745150411,
      "grad_norm": 1.1727968454360962,
      "learning_rate": 4.999215399782754e-05,
      "loss": 0.8932,
      "num_input_tokens_seen": 512224,
      "step": 320
    },
    {
      "epoch": 0.014712872631793387,
      "grad_norm": 1.9012227058410645,
      "learning_rate": 4.999190612746675e-05,
      "loss": 0.7206,
      "num_input_tokens_seen": 520000,
      "step": 325
    },
    {
      "epoch": 0.01493922451843636,
      "grad_norm": 2.5139803886413574,
      "learning_rate": 4.999165440303998e-05,
      "loss": 0.7798,
      "num_input_tokens_seen": 527616,
      "step": 330
    },
    {
      "epoch": 0.015165576405079336,
      "grad_norm": 2.0557546615600586,
      "learning_rate": 4.999139882458603e-05,
      "loss": 0.7593,
      "num_input_tokens_seen": 535776,
      "step": 335
    },
    {
      "epoch": 0.015391928291722312,
      "grad_norm": 2.5878262519836426,
      "learning_rate": 4.9991139392144314e-05,
      "loss": 0.7304,
      "num_input_tokens_seen": 543872,
      "step": 340
    },
    {
      "epoch": 0.015618280178365286,
      "grad_norm": 1.4461296796798706,
      "learning_rate": 4.999087610575485e-05,
      "loss": 0.7158,
      "num_input_tokens_seen": 551680,
      "step": 345
    },
    {
      "epoch": 0.015844632065008264,
      "grad_norm": 1.7648515701293945,
      "learning_rate": 4.999060896545824e-05,
      "loss": 0.6717,
      "num_input_tokens_seen": 559584,
      "step": 350
    },
    {
      "epoch": 0.016070983951651237,
      "grad_norm": 2.348763942718506,
      "learning_rate": 4.999033797129568e-05,
      "loss": 0.696,
      "num_input_tokens_seen": 567584,
      "step": 355
    },
    {
      "epoch": 0.01629733583829421,
      "grad_norm": 1.9555516242980957,
      "learning_rate": 4.999006312330894e-05,
      "loss": 0.6856,
      "num_input_tokens_seen": 575808,
      "step": 360
    },
    {
      "epoch": 0.01652368772493719,
      "grad_norm": 2.4099574089050293,
      "learning_rate": 4.998978442154043e-05,
      "loss": 0.5133,
      "num_input_tokens_seen": 584064,
      "step": 365
    },
    {
      "epoch": 0.016750039611580163,
      "grad_norm": 0.6025561690330505,
      "learning_rate": 4.9989501866033125e-05,
      "loss": 0.4772,
      "num_input_tokens_seen": 592160,
      "step": 370
    },
    {
      "epoch": 0.016976391498223137,
      "grad_norm": 2.713510036468506,
      "learning_rate": 4.998921545683059e-05,
      "loss": 0.7445,
      "num_input_tokens_seen": 600608,
      "step": 375
    },
    {
      "epoch": 0.017202743384866114,
      "grad_norm": 1.9176291227340698,
      "learning_rate": 4.9988925193976996e-05,
      "loss": 0.7266,
      "num_input_tokens_seen": 608544,
      "step": 380
    },
    {
      "epoch": 0.017429095271509088,
      "grad_norm": 2.3571815490722656,
      "learning_rate": 4.998863107751711e-05,
      "loss": 0.7298,
      "num_input_tokens_seen": 616032,
      "step": 385
    },
    {
      "epoch": 0.017655447158152062,
      "grad_norm": 1.8448376655578613,
      "learning_rate": 4.998833310749629e-05,
      "loss": 0.5583,
      "num_input_tokens_seen": 624800,
      "step": 390
    },
    {
      "epoch": 0.01788179904479504,
      "grad_norm": 2.0362131595611572,
      "learning_rate": 4.998803128396047e-05,
      "loss": 0.6344,
      "num_input_tokens_seen": 632768,
      "step": 395
    },
    {
      "epoch": 0.018108150931438013,
      "grad_norm": 1.6599634885787964,
      "learning_rate": 4.9987725606956215e-05,
      "loss": 0.4345,
      "num_input_tokens_seen": 640416,
      "step": 400
    },
    {
      "epoch": 0.018108150931438013,
      "eval_loss": 0.6161633133888245,
      "eval_runtime": 405.0858,
      "eval_samples_per_second": 96.945,
      "eval_steps_per_second": 24.237,
      "num_input_tokens_seen": 640416,
      "step": 400
    },
    {
      "epoch": 0.018334502818080987,
      "grad_norm": 1.534250259399414,
      "learning_rate": 4.998741607653066e-05,
      "loss": 0.5556,
      "num_input_tokens_seen": 648832,
      "step": 405
    },
    {
      "epoch": 0.018560854704723965,
      "grad_norm": 2.6987721920013428,
      "learning_rate": 4.9987102692731523e-05,
      "loss": 0.6105,
      "num_input_tokens_seen": 657600,
      "step": 410
    },
    {
      "epoch": 0.01878720659136694,
      "grad_norm": 1.6504124402999878,
      "learning_rate": 4.9986785455607157e-05,
      "loss": 0.4889,
      "num_input_tokens_seen": 665600,
      "step": 415
    },
    {
      "epoch": 0.019013558478009913,
      "grad_norm": 2.264347791671753,
      "learning_rate": 4.9986464365206456e-05,
      "loss": 0.605,
      "num_input_tokens_seen": 672960,
      "step": 420
    },
    {
      "epoch": 0.01923991036465289,
      "grad_norm": 1.9709155559539795,
      "learning_rate": 4.9986139421578956e-05,
      "loss": 0.6515,
      "num_input_tokens_seen": 680576,
      "step": 425
    },
    {
      "epoch": 0.019466262251295864,
      "grad_norm": 1.3719570636749268,
      "learning_rate": 4.998581062477477e-05,
      "loss": 0.5076,
      "num_input_tokens_seen": 688960,
      "step": 430
    },
    {
      "epoch": 0.019692614137938838,
      "grad_norm": 1.2962701320648193,
      "learning_rate": 4.998547797484458e-05,
      "loss": 0.6237,
      "num_input_tokens_seen": 697216,
      "step": 435
    },
    {
      "epoch": 0.019918966024581816,
      "grad_norm": 0.903195321559906,
      "learning_rate": 4.9985141471839706e-05,
      "loss": 0.4599,
      "num_input_tokens_seen": 705216,
      "step": 440
    },
    {
      "epoch": 0.02014531791122479,
      "grad_norm": 1.6136606931686401,
      "learning_rate": 4.998480111581203e-05,
      "loss": 0.6415,
      "num_input_tokens_seen": 713472,
      "step": 445
    },
    {
      "epoch": 0.020371669797867767,
      "grad_norm": 1.948140025138855,
      "learning_rate": 4.998445690681405e-05,
      "loss": 0.6119,
      "num_input_tokens_seen": 721792,
      "step": 450
    },
    {
      "epoch": 0.02059802168451074,
      "grad_norm": 1.2459067106246948,
      "learning_rate": 4.9984108844898834e-05,
      "loss": 0.5372,
      "num_input_tokens_seen": 729728,
      "step": 455
    },
    {
      "epoch": 0.020824373571153715,
      "grad_norm": 1.7804895639419556,
      "learning_rate": 4.9983756930120076e-05,
      "loss": 0.5746,
      "num_input_tokens_seen": 738560,
      "step": 460
    },
    {
      "epoch": 0.021050725457796692,
      "grad_norm": 1.0625650882720947,
      "learning_rate": 4.9983401162532025e-05,
      "loss": 0.7003,
      "num_input_tokens_seen": 746880,
      "step": 465
    },
    {
      "epoch": 0.021277077344439666,
      "grad_norm": 2.171696186065674,
      "learning_rate": 4.998304154218955e-05,
      "loss": 0.5104,
      "num_input_tokens_seen": 754944,
      "step": 470
    },
    {
      "epoch": 0.02150342923108264,
      "grad_norm": 1.322995662689209,
      "learning_rate": 4.998267806914812e-05,
      "loss": 0.4594,
      "num_input_tokens_seen": 762720,
      "step": 475
    },
    {
      "epoch": 0.021729781117725618,
      "grad_norm": 2.302241563796997,
      "learning_rate": 4.998231074346378e-05,
      "loss": 0.5303,
      "num_input_tokens_seen": 770720,
      "step": 480
    },
    {
      "epoch": 0.02195613300436859,
      "grad_norm": 1.5663366317749023,
      "learning_rate": 4.998193956519317e-05,
      "loss": 0.5538,
      "num_input_tokens_seen": 778592,
      "step": 485
    },
    {
      "epoch": 0.022182484891011565,
      "grad_norm": 1.0664234161376953,
      "learning_rate": 4.9981564534393545e-05,
      "loss": 0.5496,
      "num_input_tokens_seen": 787456,
      "step": 490
    },
    {
      "epoch": 0.022408836777654543,
      "grad_norm": 1.5799604654312134,
      "learning_rate": 4.998118565112272e-05,
      "loss": 0.3731,
      "num_input_tokens_seen": 795648,
      "step": 495
    },
    {
      "epoch": 0.022635188664297517,
      "grad_norm": 2.4181735515594482,
      "learning_rate": 4.998080291543914e-05,
      "loss": 0.5436,
      "num_input_tokens_seen": 803136,
      "step": 500
    },
    {
      "epoch": 0.02286154055094049,
      "grad_norm": 1.4131873846054077,
      "learning_rate": 4.9980416327401826e-05,
      "loss": 0.642,
      "num_input_tokens_seen": 811712,
      "step": 505
    },
    {
      "epoch": 0.023087892437583468,
      "grad_norm": 1.5160249471664429,
      "learning_rate": 4.998002588707038e-05,
      "loss": 0.4796,
      "num_input_tokens_seen": 820064,
      "step": 510
    },
    {
      "epoch": 0.023314244324226442,
      "grad_norm": 1.0752973556518555,
      "learning_rate": 4.997963159450503e-05,
      "loss": 0.4255,
      "num_input_tokens_seen": 828768,
      "step": 515
    },
    {
      "epoch": 0.023540596210869416,
      "grad_norm": 1.8467682600021362,
      "learning_rate": 4.9979233449766575e-05,
      "loss": 0.394,
      "num_input_tokens_seen": 836224,
      "step": 520
    },
    {
      "epoch": 0.023766948097512394,
      "grad_norm": 1.6019582748413086,
      "learning_rate": 4.997883145291641e-05,
      "loss": 0.4241,
      "num_input_tokens_seen": 844128,
      "step": 525
    },
    {
      "epoch": 0.023993299984155367,
      "grad_norm": 1.6992253065109253,
      "learning_rate": 4.9978425604016536e-05,
      "loss": 0.4463,
      "num_input_tokens_seen": 852608,
      "step": 530
    },
    {
      "epoch": 0.02421965187079834,
      "grad_norm": 1.3069809675216675,
      "learning_rate": 4.9978015903129536e-05,
      "loss": 0.4451,
      "num_input_tokens_seen": 860256,
      "step": 535
    },
    {
      "epoch": 0.02444600375744132,
      "grad_norm": 1.619842529296875,
      "learning_rate": 4.997760235031859e-05,
      "loss": 0.4313,
      "num_input_tokens_seen": 868288,
      "step": 540
    },
    {
      "epoch": 0.024672355644084293,
      "grad_norm": 1.460185170173645,
      "learning_rate": 4.9977184945647473e-05,
      "loss": 0.3566,
      "num_input_tokens_seen": 876448,
      "step": 545
    },
    {
      "epoch": 0.02489870753072727,
      "grad_norm": 1.3033572435379028,
      "learning_rate": 4.997676368918055e-05,
      "loss": 0.451,
      "num_input_tokens_seen": 884192,
      "step": 550
    },
    {
      "epoch": 0.025125059417370244,
      "grad_norm": 2.6541411876678467,
      "learning_rate": 4.9976338580982794e-05,
      "loss": 0.3174,
      "num_input_tokens_seen": 891904,
      "step": 555
    },
    {
      "epoch": 0.025351411304013218,
      "grad_norm": 1.3504552841186523,
      "learning_rate": 4.9975909621119755e-05,
      "loss": 0.4669,
      "num_input_tokens_seen": 900224,
      "step": 560
    },
    {
      "epoch": 0.025577763190656196,
      "grad_norm": 0.9813631176948547,
      "learning_rate": 4.997547680965758e-05,
      "loss": 0.3082,
      "num_input_tokens_seen": 908576,
      "step": 565
    },
    {
      "epoch": 0.02580411507729917,
      "grad_norm": 1.5656949281692505,
      "learning_rate": 4.997504014666302e-05,
      "loss": 0.3903,
      "num_input_tokens_seen": 917152,
      "step": 570
    },
    {
      "epoch": 0.026030466963942143,
      "grad_norm": 1.495326280593872,
      "learning_rate": 4.997459963220342e-05,
      "loss": 0.2766,
      "num_input_tokens_seen": 925184,
      "step": 575
    },
    {
      "epoch": 0.02625681885058512,
      "grad_norm": 1.4625545740127563,
      "learning_rate": 4.997415526634671e-05,
      "loss": 0.3258,
      "num_input_tokens_seen": 932608,
      "step": 580
    },
    {
      "epoch": 0.026483170737228095,
      "grad_norm": 1.7054921388626099,
      "learning_rate": 4.99737070491614e-05,
      "loss": 0.3732,
      "num_input_tokens_seen": 940864,
      "step": 585
    },
    {
      "epoch": 0.02670952262387107,
      "grad_norm": 1.5521199703216553,
      "learning_rate": 4.997325498071663e-05,
      "loss": 0.4672,
      "num_input_tokens_seen": 948544,
      "step": 590
    },
    {
      "epoch": 0.026935874510514046,
      "grad_norm": 0.8625586032867432,
      "learning_rate": 4.997279906108211e-05,
      "loss": 0.352,
      "num_input_tokens_seen": 956288,
      "step": 595
    },
    {
      "epoch": 0.02716222639715702,
      "grad_norm": 1.3800281286239624,
      "learning_rate": 4.9972339290328155e-05,
      "loss": 0.3422,
      "num_input_tokens_seen": 964160,
      "step": 600
    },
    {
      "epoch": 0.02716222639715702,
      "eval_loss": 0.37266644835472107,
      "eval_runtime": 404.3581,
      "eval_samples_per_second": 97.119,
      "eval_steps_per_second": 24.28,
      "num_input_tokens_seen": 964160,
      "step": 600
    },
    {
      "epoch": 0.027388578283799994,
      "grad_norm": 0.9569019079208374,
      "learning_rate": 4.9971875668525646e-05,
      "loss": 0.3387,
      "num_input_tokens_seen": 971872,
      "step": 605
    },
    {
      "epoch": 0.02761493017044297,
      "grad_norm": 1.3284109830856323,
      "learning_rate": 4.997140819574609e-05,
      "loss": 0.4051,
      "num_input_tokens_seen": 979360,
      "step": 610
    },
    {
      "epoch": 0.027841282057085946,
      "grad_norm": 0.6574214696884155,
      "learning_rate": 4.997093687206159e-05,
      "loss": 0.3785,
      "num_input_tokens_seen": 986944,
      "step": 615
    },
    {
      "epoch": 0.02806763394372892,
      "grad_norm": 0.8225143551826477,
      "learning_rate": 4.997046169754482e-05,
      "loss": 0.3668,
      "num_input_tokens_seen": 995264,
      "step": 620
    },
    {
      "epoch": 0.028293985830371897,
      "grad_norm": 0.9075181484222412,
      "learning_rate": 4.996998267226905e-05,
      "loss": 0.3227,
      "num_input_tokens_seen": 1003072,
      "step": 625
    },
    {
      "epoch": 0.02852033771701487,
      "grad_norm": 1.427632212638855,
      "learning_rate": 4.996949979630817e-05,
      "loss": 0.4046,
      "num_input_tokens_seen": 1011584,
      "step": 630
    },
    {
      "epoch": 0.028746689603657845,
      "grad_norm": 1.6144381761550903,
      "learning_rate": 4.996901306973663e-05,
      "loss": 0.287,
      "num_input_tokens_seen": 1019520,
      "step": 635
    },
    {
      "epoch": 0.028973041490300822,
      "grad_norm": 2.5626771450042725,
      "learning_rate": 4.996852249262949e-05,
      "loss": 0.4088,
      "num_input_tokens_seen": 1027200,
      "step": 640
    },
    {
      "epoch": 0.029199393376943796,
      "grad_norm": 1.0725980997085571,
      "learning_rate": 4.996802806506241e-05,
      "loss": 0.261,
      "num_input_tokens_seen": 1034944,
      "step": 645
    },
    {
      "epoch": 0.029425745263586774,
      "grad_norm": 0.9125100374221802,
      "learning_rate": 4.996752978711164e-05,
      "loss": 0.3251,
      "num_input_tokens_seen": 1043264,
      "step": 650
    },
    {
      "epoch": 0.029652097150229748,
      "grad_norm": 0.9181780815124512,
      "learning_rate": 4.996702765885401e-05,
      "loss": 0.3668,
      "num_input_tokens_seen": 1051456,
      "step": 655
    },
    {
      "epoch": 0.02987844903687272,
      "grad_norm": 1.115751028060913,
      "learning_rate": 4.9966521680366964e-05,
      "loss": 0.4506,
      "num_input_tokens_seen": 1059360,
      "step": 660
    },
    {
      "epoch": 0.0301048009235157,
      "grad_norm": 1.296939492225647,
      "learning_rate": 4.9966011851728524e-05,
      "loss": 0.2828,
      "num_input_tokens_seen": 1067552,
      "step": 665
    },
    {
      "epoch": 0.030331152810158673,
      "grad_norm": 0.8733476996421814,
      "learning_rate": 4.996549817301731e-05,
      "loss": 0.3191,
      "num_input_tokens_seen": 1075232,
      "step": 670
    },
    {
      "epoch": 0.030557504696801647,
      "grad_norm": 1.1066807508468628,
      "learning_rate": 4.9964980644312544e-05,
      "loss": 0.2891,
      "num_input_tokens_seen": 1084160,
      "step": 675
    },
    {
      "epoch": 0.030783856583444624,
      "grad_norm": 1.074399471282959,
      "learning_rate": 4.996445926569403e-05,
      "loss": 0.3556,
      "num_input_tokens_seen": 1092096,
      "step": 680
    },
    {
      "epoch": 0.031010208470087598,
      "grad_norm": 1.2012834548950195,
      "learning_rate": 4.996393403724218e-05,
      "loss": 0.361,
      "num_input_tokens_seen": 1099648,
      "step": 685
    },
    {
      "epoch": 0.031236560356730572,
      "grad_norm": 1.3543130159378052,
      "learning_rate": 4.9963404959037985e-05,
      "loss": 0.2993,
      "num_input_tokens_seen": 1107840,
      "step": 690
    },
    {
      "epoch": 0.031462912243373546,
      "grad_norm": 0.9810236096382141,
      "learning_rate": 4.996287203116303e-05,
      "loss": 0.2709,
      "num_input_tokens_seen": 1115424,
      "step": 695
    },
    {
      "epoch": 0.03168926413001653,
      "grad_norm": 0.8294551968574524,
      "learning_rate": 4.996233525369951e-05,
      "loss": 0.2646,
      "num_input_tokens_seen": 1123392,
      "step": 700
    },
    {
      "epoch": 0.0319156160166595,
      "grad_norm": 1.1061280965805054,
      "learning_rate": 4.99617946267302e-05,
      "loss": 0.3297,
      "num_input_tokens_seen": 1131744,
      "step": 705
    },
    {
      "epoch": 0.032141967903302475,
      "grad_norm": 0.6570104956626892,
      "learning_rate": 4.996125015033846e-05,
      "loss": 0.3313,
      "num_input_tokens_seen": 1139200,
      "step": 710
    },
    {
      "epoch": 0.03236831978994545,
      "grad_norm": 0.9010317921638489,
      "learning_rate": 4.996070182460827e-05,
      "loss": 0.361,
      "num_input_tokens_seen": 1146848,
      "step": 715
    },
    {
      "epoch": 0.03259467167658842,
      "grad_norm": 1.405372142791748,
      "learning_rate": 4.996014964962418e-05,
      "loss": 0.3516,
      "num_input_tokens_seen": 1155104,
      "step": 720
    },
    {
      "epoch": 0.0328210235632314,
      "grad_norm": 0.56741863489151,
      "learning_rate": 4.9959593625471344e-05,
      "loss": 0.2643,
      "num_input_tokens_seen": 1163168,
      "step": 725
    },
    {
      "epoch": 0.03304737544987438,
      "grad_norm": 1.0027179718017578,
      "learning_rate": 4.995903375223552e-05,
      "loss": 0.3173,
      "num_input_tokens_seen": 1171200,
      "step": 730
    },
    {
      "epoch": 0.03327372733651735,
      "grad_norm": 1.2155508995056152,
      "learning_rate": 4.995847003000302e-05,
      "loss": 0.3706,
      "num_input_tokens_seen": 1179680,
      "step": 735
    },
    {
      "epoch": 0.033500079223160326,
      "grad_norm": 1.0001428127288818,
      "learning_rate": 4.9957902458860804e-05,
      "loss": 0.3073,
      "num_input_tokens_seen": 1187520,
      "step": 740
    },
    {
      "epoch": 0.0337264311098033,
      "grad_norm": 1.9576117992401123,
      "learning_rate": 4.995733103889639e-05,
      "loss": 0.3775,
      "num_input_tokens_seen": 1195264,
      "step": 745
    },
    {
      "epoch": 0.033952782996446274,
      "grad_norm": 0.7967370748519897,
      "learning_rate": 4.99567557701979e-05,
      "loss": 0.2741,
      "num_input_tokens_seen": 1203648,
      "step": 750
    },
    {
      "epoch": 0.03417913488308925,
      "grad_norm": 1.1833633184432983,
      "learning_rate": 4.995617665285403e-05,
      "loss": 0.2601,
      "num_input_tokens_seen": 1212032,
      "step": 755
    },
    {
      "epoch": 0.03440548676973223,
      "grad_norm": 0.5846611857414246,
      "learning_rate": 4.99555936869541e-05,
      "loss": 0.2361,
      "num_input_tokens_seen": 1219552,
      "step": 760
    },
    {
      "epoch": 0.0346318386563752,
      "grad_norm": 1.0174949169158936,
      "learning_rate": 4.995500687258803e-05,
      "loss": 0.2945,
      "num_input_tokens_seen": 1227520,
      "step": 765
    },
    {
      "epoch": 0.034858190543018176,
      "grad_norm": 1.2132169008255005,
      "learning_rate": 4.995441620984628e-05,
      "loss": 0.3533,
      "num_input_tokens_seen": 1235264,
      "step": 770
    },
    {
      "epoch": 0.03508454242966115,
      "grad_norm": 0.8293858170509338,
      "learning_rate": 4.995382169881996e-05,
      "loss": 0.2514,
      "num_input_tokens_seen": 1243456,
      "step": 775
    },
    {
      "epoch": 0.035310894316304124,
      "grad_norm": 0.768282949924469,
      "learning_rate": 4.9953223339600755e-05,
      "loss": 0.2615,
      "num_input_tokens_seen": 1251872,
      "step": 780
    },
    {
      "epoch": 0.0355372462029471,
      "grad_norm": 0.8564805388450623,
      "learning_rate": 4.995262113228091e-05,
      "loss": 0.2693,
      "num_input_tokens_seen": 1260064,
      "step": 785
    },
    {
      "epoch": 0.03576359808959008,
      "grad_norm": 0.6368021368980408,
      "learning_rate": 4.995201507695332e-05,
      "loss": 0.3225,
      "num_input_tokens_seen": 1268480,
      "step": 790
    },
    {
      "epoch": 0.03598994997623305,
      "grad_norm": 0.6616157293319702,
      "learning_rate": 4.995140517371144e-05,
      "loss": 0.2844,
      "num_input_tokens_seen": 1276736,
      "step": 795
    },
    {
      "epoch": 0.03621630186287603,
      "grad_norm": 1.1118320226669312,
      "learning_rate": 4.995079142264932e-05,
      "loss": 0.393,
      "num_input_tokens_seen": 1285344,
      "step": 800
    },
    {
      "epoch": 0.03621630186287603,
      "eval_loss": 0.28129151463508606,
      "eval_runtime": 404.5124,
      "eval_samples_per_second": 97.082,
      "eval_steps_per_second": 24.271,
      "num_input_tokens_seen": 1285344,
      "step": 800
    },
    {
      "epoch": 0.036442653749519,
      "grad_norm": 0.8840278387069702,
      "learning_rate": 4.995017382386162e-05,
      "loss": 0.2975,
      "num_input_tokens_seen": 1293280,
      "step": 805
    },
    {
      "epoch": 0.036669005636161975,
      "grad_norm": 0.8421594500541687,
      "learning_rate": 4.994955237744356e-05,
      "loss": 0.3442,
      "num_input_tokens_seen": 1301280,
      "step": 810
    },
    {
      "epoch": 0.036895357522804956,
      "grad_norm": 0.48256608843803406,
      "learning_rate": 4.994892708349101e-05,
      "loss": 0.2636,
      "num_input_tokens_seen": 1309216,
      "step": 815
    },
    {
      "epoch": 0.03712170940944793,
      "grad_norm": 0.8255535960197449,
      "learning_rate": 4.994829794210035e-05,
      "loss": 0.237,
      "num_input_tokens_seen": 1317344,
      "step": 820
    },
    {
      "epoch": 0.037348061296090904,
      "grad_norm": 0.9009722471237183,
      "learning_rate": 4.994766495336864e-05,
      "loss": 0.277,
      "num_input_tokens_seen": 1325024,
      "step": 825
    },
    {
      "epoch": 0.03757441318273388,
      "grad_norm": 0.7371839880943298,
      "learning_rate": 4.994702811739348e-05,
      "loss": 0.2571,
      "num_input_tokens_seen": 1333280,
      "step": 830
    },
    {
      "epoch": 0.03780076506937685,
      "grad_norm": 0.9540840983390808,
      "learning_rate": 4.994638743427308e-05,
      "loss": 0.3042,
      "num_input_tokens_seen": 1341216,
      "step": 835
    },
    {
      "epoch": 0.038027116956019825,
      "grad_norm": 1.4647727012634277,
      "learning_rate": 4.994574290410624e-05,
      "loss": 0.3764,
      "num_input_tokens_seen": 1350112,
      "step": 840
    },
    {
      "epoch": 0.038253468842662806,
      "grad_norm": 0.798035740852356,
      "learning_rate": 4.9945094526992364e-05,
      "loss": 0.3238,
      "num_input_tokens_seen": 1359104,
      "step": 845
    },
    {
      "epoch": 0.03847982072930578,
      "grad_norm": 0.6058248281478882,
      "learning_rate": 4.994444230303142e-05,
      "loss": 0.2384,
      "num_input_tokens_seen": 1366816,
      "step": 850
    },
    {
      "epoch": 0.038706172615948754,
      "grad_norm": 0.9413124918937683,
      "learning_rate": 4.994378623232402e-05,
      "loss": 0.2267,
      "num_input_tokens_seen": 1374528,
      "step": 855
    },
    {
      "epoch": 0.03893252450259173,
      "grad_norm": 0.5582336187362671,
      "learning_rate": 4.99431263149713e-05,
      "loss": 0.2746,
      "num_input_tokens_seen": 1382368,
      "step": 860
    },
    {
      "epoch": 0.0391588763892347,
      "grad_norm": 0.7686482071876526,
      "learning_rate": 4.9942462551075056e-05,
      "loss": 0.2231,
      "num_input_tokens_seen": 1389888,
      "step": 865
    },
    {
      "epoch": 0.039385228275877676,
      "grad_norm": 0.8412222266197205,
      "learning_rate": 4.994179494073764e-05,
      "loss": 0.2498,
      "num_input_tokens_seen": 1398176,
      "step": 870
    },
    {
      "epoch": 0.03961158016252066,
      "grad_norm": 1.1730015277862549,
      "learning_rate": 4.9941123484062e-05,
      "loss": 0.2787,
      "num_input_tokens_seen": 1406208,
      "step": 875
    },
    {
      "epoch": 0.03983793204916363,
      "grad_norm": 0.848338782787323,
      "learning_rate": 4.99404481811517e-05,
      "loss": 0.3443,
      "num_input_tokens_seen": 1414720,
      "step": 880
    },
    {
      "epoch": 0.040064283935806605,
      "grad_norm": 0.5007281303405762,
      "learning_rate": 4.9939769032110864e-05,
      "loss": 0.2802,
      "num_input_tokens_seen": 1422304,
      "step": 885
    },
    {
      "epoch": 0.04029063582244958,
      "grad_norm": 0.7899531722068787,
      "learning_rate": 4.993908603704423e-05,
      "loss": 0.2679,
      "num_input_tokens_seen": 1430528,
      "step": 890
    },
    {
      "epoch": 0.04051698770909255,
      "grad_norm": 0.695751965045929,
      "learning_rate": 4.9938399196057126e-05,
      "loss": 0.3194,
      "num_input_tokens_seen": 1438720,
      "step": 895
    },
    {
      "epoch": 0.040743339595735534,
      "grad_norm": 0.8114869594573975,
      "learning_rate": 4.993770850925547e-05,
      "loss": 0.2587,
      "num_input_tokens_seen": 1446528,
      "step": 900
    },
    {
      "epoch": 0.04096969148237851,
      "grad_norm": 0.9676947593688965,
      "learning_rate": 4.993701397674577e-05,
      "loss": 0.2291,
      "num_input_tokens_seen": 1454432,
      "step": 905
    },
    {
      "epoch": 0.04119604336902148,
      "grad_norm": 0.7571606040000916,
      "learning_rate": 4.993631559863515e-05,
      "loss": 0.2484,
      "num_input_tokens_seen": 1463136,
      "step": 910
    },
    {
      "epoch": 0.041422395255664456,
      "grad_norm": 0.5244556069374084,
      "learning_rate": 4.9935613375031283e-05,
      "loss": 0.2478,
      "num_input_tokens_seen": 1471360,
      "step": 915
    },
    {
      "epoch": 0.04164874714230743,
      "grad_norm": 0.39562860131263733,
      "learning_rate": 4.993490730604248e-05,
      "loss": 0.2697,
      "num_input_tokens_seen": 1479296,
      "step": 920
    },
    {
      "epoch": 0.041875099028950404,
      "grad_norm": 0.8263768553733826,
      "learning_rate": 4.993419739177761e-05,
      "loss": 0.2776,
      "num_input_tokens_seen": 1487552,
      "step": 925
    },
    {
      "epoch": 0.042101450915593384,
      "grad_norm": 0.7280258536338806,
      "learning_rate": 4.9933483632346164e-05,
      "loss": 0.2665,
      "num_input_tokens_seen": 1495424,
      "step": 930
    },
    {
      "epoch": 0.04232780280223636,
      "grad_norm": 1.0047080516815186,
      "learning_rate": 4.993276602785821e-05,
      "loss": 0.2123,
      "num_input_tokens_seen": 1503520,
      "step": 935
    },
    {
      "epoch": 0.04255415468887933,
      "grad_norm": 0.503893256187439,
      "learning_rate": 4.993204457842441e-05,
      "loss": 0.2142,
      "num_input_tokens_seen": 1511200,
      "step": 940
    },
    {
      "epoch": 0.042780506575522306,
      "grad_norm": 0.6510166525840759,
      "learning_rate": 4.993131928415602e-05,
      "loss": 0.2618,
      "num_input_tokens_seen": 1519168,
      "step": 945
    },
    {
      "epoch": 0.04300685846216528,
      "grad_norm": 0.9346144795417786,
      "learning_rate": 4.993059014516489e-05,
      "loss": 0.2509,
      "num_input_tokens_seen": 1527296,
      "step": 950
    },
    {
      "epoch": 0.043233210348808254,
      "grad_norm": 0.4940048158168793,
      "learning_rate": 4.9929857161563464e-05,
      "loss": 0.2757,
      "num_input_tokens_seen": 1535296,
      "step": 955
    },
    {
      "epoch": 0.043459562235451235,
      "grad_norm": 0.5405433177947998,
      "learning_rate": 4.992912033346477e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 1543520,
      "step": 960
    },
    {
      "epoch": 0.04368591412209421,
      "grad_norm": 0.7343841791152954,
      "learning_rate": 4.992837966098245e-05,
      "loss": 0.2585,
      "num_input_tokens_seen": 1551520,
      "step": 965
    },
    {
      "epoch": 0.04391226600873718,
      "grad_norm": 0.5691529512405396,
      "learning_rate": 4.992763514423071e-05,
      "loss": 0.235,
      "num_input_tokens_seen": 1560416,
      "step": 970
    },
    {
      "epoch": 0.04413861789538016,
      "grad_norm": 1.2914742231369019,
      "learning_rate": 4.992688678332437e-05,
      "loss": 0.2587,
      "num_input_tokens_seen": 1568128,
      "step": 975
    },
    {
      "epoch": 0.04436496978202313,
      "grad_norm": 0.7327566742897034,
      "learning_rate": 4.992613457837884e-05,
      "loss": 0.2561,
      "num_input_tokens_seen": 1576416,
      "step": 980
    },
    {
      "epoch": 0.04459132166866611,
      "grad_norm": 0.5650726556777954,
      "learning_rate": 4.992537852951011e-05,
      "loss": 0.2349,
      "num_input_tokens_seen": 1584768,
      "step": 985
    },
    {
      "epoch": 0.044817673555309086,
      "grad_norm": 0.6310359239578247,
      "learning_rate": 4.9924618636834785e-05,
      "loss": 0.2533,
      "num_input_tokens_seen": 1592576,
      "step": 990
    },
    {
      "epoch": 0.04504402544195206,
      "grad_norm": 1.012549877166748,
      "learning_rate": 4.9923854900470046e-05,
      "loss": 0.2858,
      "num_input_tokens_seen": 1600544,
      "step": 995
    },
    {
      "epoch": 0.045270377328595034,
      "grad_norm": 0.8358775973320007,
      "learning_rate": 4.992308732053367e-05,
      "loss": 0.2407,
      "num_input_tokens_seen": 1608256,
      "step": 1000
    },
    {
      "epoch": 0.045270377328595034,
      "eval_loss": 0.24932870268821716,
      "eval_runtime": 404.6857,
      "eval_samples_per_second": 97.041,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 1608256,
      "step": 1000
    },
    {
      "epoch": 0.04549672921523801,
      "grad_norm": 1.3549906015396118,
      "learning_rate": 4.992231589714402e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 1616384,
      "step": 1005
    },
    {
      "epoch": 0.04572308110188098,
      "grad_norm": 0.5121874809265137,
      "learning_rate": 4.992154063042007e-05,
      "loss": 0.2537,
      "num_input_tokens_seen": 1623872,
      "step": 1010
    },
    {
      "epoch": 0.04594943298852396,
      "grad_norm": 0.6890076994895935,
      "learning_rate": 4.992076152048136e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 1631936,
      "step": 1015
    },
    {
      "epoch": 0.046175784875166936,
      "grad_norm": 0.34070444107055664,
      "learning_rate": 4.991997856744807e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 1640192,
      "step": 1020
    },
    {
      "epoch": 0.04640213676180991,
      "grad_norm": 0.36056244373321533,
      "learning_rate": 4.9919191771440905e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 1648288,
      "step": 1025
    },
    {
      "epoch": 0.046628488648452884,
      "grad_norm": 1.134379506111145,
      "learning_rate": 4.991840113258122e-05,
      "loss": 0.2679,
      "num_input_tokens_seen": 1657312,
      "step": 1030
    },
    {
      "epoch": 0.04685484053509586,
      "grad_norm": 0.5258514881134033,
      "learning_rate": 4.9917606650990933e-05,
      "loss": 0.3083,
      "num_input_tokens_seen": 1664800,
      "step": 1035
    },
    {
      "epoch": 0.04708119242173883,
      "grad_norm": 0.6280168294906616,
      "learning_rate": 4.9916808326792566e-05,
      "loss": 0.2511,
      "num_input_tokens_seen": 1673184,
      "step": 1040
    },
    {
      "epoch": 0.04730754430838181,
      "grad_norm": 0.4857400059700012,
      "learning_rate": 4.9916006160109235e-05,
      "loss": 0.2235,
      "num_input_tokens_seen": 1680928,
      "step": 1045
    },
    {
      "epoch": 0.04753389619502479,
      "grad_norm": 0.7146949768066406,
      "learning_rate": 4.991520015106464e-05,
      "loss": 0.2523,
      "num_input_tokens_seen": 1688928,
      "step": 1050
    },
    {
      "epoch": 0.04776024808166776,
      "grad_norm": 0.526668131351471,
      "learning_rate": 4.991439029978308e-05,
      "loss": 0.3242,
      "num_input_tokens_seen": 1697024,
      "step": 1055
    },
    {
      "epoch": 0.047986599968310735,
      "grad_norm": 0.7500360012054443,
      "learning_rate": 4.9913576606389434e-05,
      "loss": 0.2573,
      "num_input_tokens_seen": 1704992,
      "step": 1060
    },
    {
      "epoch": 0.04821295185495371,
      "grad_norm": 0.771552562713623,
      "learning_rate": 4.991275907100919e-05,
      "loss": 0.2451,
      "num_input_tokens_seen": 1712640,
      "step": 1065
    },
    {
      "epoch": 0.04843930374159668,
      "grad_norm": 0.7034665942192078,
      "learning_rate": 4.9911937693768434e-05,
      "loss": 0.2114,
      "num_input_tokens_seen": 1721216,
      "step": 1070
    },
    {
      "epoch": 0.048665655628239664,
      "grad_norm": 1.198737621307373,
      "learning_rate": 4.991111247479382e-05,
      "loss": 0.2092,
      "num_input_tokens_seen": 1729504,
      "step": 1075
    },
    {
      "epoch": 0.04889200751488264,
      "grad_norm": 0.95503169298172,
      "learning_rate": 4.9910283414212605e-05,
      "loss": 0.265,
      "num_input_tokens_seen": 1736928,
      "step": 1080
    },
    {
      "epoch": 0.04911835940152561,
      "grad_norm": 0.547565758228302,
      "learning_rate": 4.990945051215265e-05,
      "loss": 0.2015,
      "num_input_tokens_seen": 1744768,
      "step": 1085
    },
    {
      "epoch": 0.049344711288168586,
      "grad_norm": 0.8849241137504578,
      "learning_rate": 4.99086137687424e-05,
      "loss": 0.2402,
      "num_input_tokens_seen": 1752480,
      "step": 1090
    },
    {
      "epoch": 0.04957106317481156,
      "grad_norm": 1.2272812128067017,
      "learning_rate": 4.9907773184110874e-05,
      "loss": 0.2405,
      "num_input_tokens_seen": 1760416,
      "step": 1095
    },
    {
      "epoch": 0.04979741506145454,
      "grad_norm": 0.6098403334617615,
      "learning_rate": 4.9906928758387715e-05,
      "loss": 0.226,
      "num_input_tokens_seen": 1768192,
      "step": 1100
    },
    {
      "epoch": 0.050023766948097514,
      "grad_norm": 1.0415542125701904,
      "learning_rate": 4.9906080491703146e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 1776736,
      "step": 1105
    },
    {
      "epoch": 0.05025011883474049,
      "grad_norm": 0.8220102787017822,
      "learning_rate": 4.990522838418797e-05,
      "loss": 0.2161,
      "num_input_tokens_seen": 1785856,
      "step": 1110
    },
    {
      "epoch": 0.05047647072138346,
      "grad_norm": 0.5358800888061523,
      "learning_rate": 4.9904372435973604e-05,
      "loss": 0.2352,
      "num_input_tokens_seen": 1794368,
      "step": 1115
    },
    {
      "epoch": 0.050702822608026436,
      "grad_norm": 0.7343965172767639,
      "learning_rate": 4.990351264719203e-05,
      "loss": 0.2805,
      "num_input_tokens_seen": 1802912,
      "step": 1120
    },
    {
      "epoch": 0.05092917449466941,
      "grad_norm": 0.43716976046562195,
      "learning_rate": 4.990264901797586e-05,
      "loss": 0.238,
      "num_input_tokens_seen": 1811264,
      "step": 1125
    },
    {
      "epoch": 0.05115552638131239,
      "grad_norm": 0.8945620059967041,
      "learning_rate": 4.990178154845826e-05,
      "loss": 0.2066,
      "num_input_tokens_seen": 1819840,
      "step": 1130
    },
    {
      "epoch": 0.051381878267955365,
      "grad_norm": 0.8375337719917297,
      "learning_rate": 4.9900910238773014e-05,
      "loss": 0.2702,
      "num_input_tokens_seen": 1827552,
      "step": 1135
    },
    {
      "epoch": 0.05160823015459834,
      "grad_norm": 0.5019015669822693,
      "learning_rate": 4.990003508905448e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 1835840,
      "step": 1140
    },
    {
      "epoch": 0.05183458204124131,
      "grad_norm": 0.3983471393585205,
      "learning_rate": 4.989915609943763e-05,
      "loss": 0.2629,
      "num_input_tokens_seen": 1844000,
      "step": 1145
    },
    {
      "epoch": 0.05206093392788429,
      "grad_norm": 1.0761064291000366,
      "learning_rate": 4.9898273270058e-05,
      "loss": 0.2262,
      "num_input_tokens_seen": 1851648,
      "step": 1150
    },
    {
      "epoch": 0.05228728581452726,
      "grad_norm": 1.6153485774993896,
      "learning_rate": 4.989738660105174e-05,
      "loss": 0.3267,
      "num_input_tokens_seen": 1859872,
      "step": 1155
    },
    {
      "epoch": 0.05251363770117024,
      "grad_norm": 0.4249184727668762,
      "learning_rate": 4.989649609255559e-05,
      "loss": 0.2313,
      "num_input_tokens_seen": 1868096,
      "step": 1160
    },
    {
      "epoch": 0.052739989587813216,
      "grad_norm": 0.7358641624450684,
      "learning_rate": 4.989560174470687e-05,
      "loss": 0.2237,
      "num_input_tokens_seen": 1875648,
      "step": 1165
    },
    {
      "epoch": 0.05296634147445619,
      "grad_norm": 0.7078103423118591,
      "learning_rate": 4.989470355764351e-05,
      "loss": 0.2508,
      "num_input_tokens_seen": 1883136,
      "step": 1170
    },
    {
      "epoch": 0.053192693361099164,
      "grad_norm": 0.7861230969429016,
      "learning_rate": 4.9893801531504e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 1891392,
      "step": 1175
    },
    {
      "epoch": 0.05341904524774214,
      "grad_norm": 0.49716320633888245,
      "learning_rate": 4.9892895666427475e-05,
      "loss": 0.2051,
      "num_input_tokens_seen": 1899552,
      "step": 1180
    },
    {
      "epoch": 0.05364539713438512,
      "grad_norm": 0.44628772139549255,
      "learning_rate": 4.9891985962553606e-05,
      "loss": 0.2685,
      "num_input_tokens_seen": 1907392,
      "step": 1185
    },
    {
      "epoch": 0.05387174902102809,
      "grad_norm": 0.429994136095047,
      "learning_rate": 4.989107242002269e-05,
      "loss": 0.2333,
      "num_input_tokens_seen": 1915680,
      "step": 1190
    },
    {
      "epoch": 0.054098100907671066,
      "grad_norm": 1.2088040113449097,
      "learning_rate": 4.989015503897561e-05,
      "loss": 0.2522,
      "num_input_tokens_seen": 1923680,
      "step": 1195
    },
    {
      "epoch": 0.05432445279431404,
      "grad_norm": 1.3861538171768188,
      "learning_rate": 4.988923381955383e-05,
      "loss": 0.3042,
      "num_input_tokens_seen": 1932352,
      "step": 1200
    },
    {
      "epoch": 0.05432445279431404,
      "eval_loss": 0.23525065183639526,
      "eval_runtime": 404.6228,
      "eval_samples_per_second": 97.056,
      "eval_steps_per_second": 24.265,
      "num_input_tokens_seen": 1932352,
      "step": 1200
    },
    {
      "epoch": 0.054550804680957014,
      "grad_norm": 0.7452578544616699,
      "learning_rate": 4.988830876189942e-05,
      "loss": 0.2709,
      "num_input_tokens_seen": 1939904,
      "step": 1205
    },
    {
      "epoch": 0.05477715656759999,
      "grad_norm": 0.8113817572593689,
      "learning_rate": 4.988737986615503e-05,
      "loss": 0.2856,
      "num_input_tokens_seen": 1947872,
      "step": 1210
    },
    {
      "epoch": 0.05500350845424297,
      "grad_norm": 0.6792585253715515,
      "learning_rate": 4.988644713246391e-05,
      "loss": 0.2276,
      "num_input_tokens_seen": 1955776,
      "step": 1215
    },
    {
      "epoch": 0.05522986034088594,
      "grad_norm": 0.42102715373039246,
      "learning_rate": 4.988551056096991e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 1963808,
      "step": 1220
    },
    {
      "epoch": 0.05545621222752892,
      "grad_norm": 0.5339102745056152,
      "learning_rate": 4.988457015181743e-05,
      "loss": 0.2185,
      "num_input_tokens_seen": 1972384,
      "step": 1225
    },
    {
      "epoch": 0.05568256411417189,
      "grad_norm": 1.051864504814148,
      "learning_rate": 4.988362590515153e-05,
      "loss": 0.2527,
      "num_input_tokens_seen": 1980800,
      "step": 1230
    },
    {
      "epoch": 0.055908916000814865,
      "grad_norm": 0.3510971963405609,
      "learning_rate": 4.9882677821117805e-05,
      "loss": 0.2429,
      "num_input_tokens_seen": 1988960,
      "step": 1235
    },
    {
      "epoch": 0.05613526788745784,
      "grad_norm": 0.8727486729621887,
      "learning_rate": 4.988172589986246e-05,
      "loss": 0.2646,
      "num_input_tokens_seen": 1996608,
      "step": 1240
    },
    {
      "epoch": 0.05636161977410082,
      "grad_norm": 0.558040976524353,
      "learning_rate": 4.9880770141532304e-05,
      "loss": 0.2391,
      "num_input_tokens_seen": 2004992,
      "step": 1245
    },
    {
      "epoch": 0.056587971660743794,
      "grad_norm": 0.5784221291542053,
      "learning_rate": 4.987981054627472e-05,
      "loss": 0.259,
      "num_input_tokens_seen": 2013056,
      "step": 1250
    },
    {
      "epoch": 0.05681432354738677,
      "grad_norm": 0.7153916954994202,
      "learning_rate": 4.987884711423769e-05,
      "loss": 0.2561,
      "num_input_tokens_seen": 2021088,
      "step": 1255
    },
    {
      "epoch": 0.05704067543402974,
      "grad_norm": 1.109047532081604,
      "learning_rate": 4.9877879845569784e-05,
      "loss": 0.2179,
      "num_input_tokens_seen": 2029536,
      "step": 1260
    },
    {
      "epoch": 0.057267027320672716,
      "grad_norm": 0.8732477426528931,
      "learning_rate": 4.9876908740420175e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 2037312,
      "step": 1265
    },
    {
      "epoch": 0.05749337920731569,
      "grad_norm": 0.6069647073745728,
      "learning_rate": 4.987593379893861e-05,
      "loss": 0.2226,
      "num_input_tokens_seen": 2045056,
      "step": 1270
    },
    {
      "epoch": 0.05771973109395867,
      "grad_norm": 0.9720463752746582,
      "learning_rate": 4.987495502127545e-05,
      "loss": 0.2426,
      "num_input_tokens_seen": 2053248,
      "step": 1275
    },
    {
      "epoch": 0.057946082980601644,
      "grad_norm": 0.5892982482910156,
      "learning_rate": 4.987397240758162e-05,
      "loss": 0.2691,
      "num_input_tokens_seen": 2061760,
      "step": 1280
    },
    {
      "epoch": 0.05817243486724462,
      "grad_norm": 0.9166043996810913,
      "learning_rate": 4.9872985958008664e-05,
      "loss": 0.2344,
      "num_input_tokens_seen": 2070336,
      "step": 1285
    },
    {
      "epoch": 0.05839878675388759,
      "grad_norm": 0.618552565574646,
      "learning_rate": 4.987199567270871e-05,
      "loss": 0.2022,
      "num_input_tokens_seen": 2078336,
      "step": 1290
    },
    {
      "epoch": 0.058625138640530566,
      "grad_norm": 0.42869627475738525,
      "learning_rate": 4.9871001551834444e-05,
      "loss": 0.2614,
      "num_input_tokens_seen": 2086176,
      "step": 1295
    },
    {
      "epoch": 0.05885149052717355,
      "grad_norm": 0.5739874839782715,
      "learning_rate": 4.98700035955392e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 2094400,
      "step": 1300
    },
    {
      "epoch": 0.05907784241381652,
      "grad_norm": 1.2421557903289795,
      "learning_rate": 4.986900180397686e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 2102016,
      "step": 1305
    },
    {
      "epoch": 0.059304194300459495,
      "grad_norm": 0.5432956218719482,
      "learning_rate": 4.9867996177301926e-05,
      "loss": 0.202,
      "num_input_tokens_seen": 2110048,
      "step": 1310
    },
    {
      "epoch": 0.05953054618710247,
      "grad_norm": 0.8526095747947693,
      "learning_rate": 4.9866986715669464e-05,
      "loss": 0.2124,
      "num_input_tokens_seen": 2117856,
      "step": 1315
    },
    {
      "epoch": 0.05975689807374544,
      "grad_norm": 0.5082677602767944,
      "learning_rate": 4.9865973419235155e-05,
      "loss": 0.224,
      "num_input_tokens_seen": 2125984,
      "step": 1320
    },
    {
      "epoch": 0.05998324996038842,
      "grad_norm": 0.5195935368537903,
      "learning_rate": 4.986495628815526e-05,
      "loss": 0.2379,
      "num_input_tokens_seen": 2133632,
      "step": 1325
    },
    {
      "epoch": 0.0602096018470314,
      "grad_norm": 1.080201268196106,
      "learning_rate": 4.986393532258663e-05,
      "loss": 0.275,
      "num_input_tokens_seen": 2141440,
      "step": 1330
    },
    {
      "epoch": 0.06043595373367437,
      "grad_norm": 0.660354495048523,
      "learning_rate": 4.986291052268671e-05,
      "loss": 0.194,
      "num_input_tokens_seen": 2149312,
      "step": 1335
    },
    {
      "epoch": 0.060662305620317346,
      "grad_norm": 0.8874869346618652,
      "learning_rate": 4.986188188861355e-05,
      "loss": 0.2299,
      "num_input_tokens_seen": 2157472,
      "step": 1340
    },
    {
      "epoch": 0.06088865750696032,
      "grad_norm": 0.5256014466285706,
      "learning_rate": 4.9860849420525766e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 2165088,
      "step": 1345
    },
    {
      "epoch": 0.061115009393603294,
      "grad_norm": 0.639656126499176,
      "learning_rate": 4.9859813118582575e-05,
      "loss": 0.2425,
      "num_input_tokens_seen": 2173088,
      "step": 1350
    },
    {
      "epoch": 0.06134136128024627,
      "grad_norm": 0.39145535230636597,
      "learning_rate": 4.98587729829438e-05,
      "loss": 0.2335,
      "num_input_tokens_seen": 2180768,
      "step": 1355
    },
    {
      "epoch": 0.06156771316688925,
      "grad_norm": 0.36414459347724915,
      "learning_rate": 4.985772901376983e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 2189152,
      "step": 1360
    },
    {
      "epoch": 0.06179406505353222,
      "grad_norm": 0.8490943312644958,
      "learning_rate": 4.9856681211221666e-05,
      "loss": 0.2203,
      "num_input_tokens_seen": 2197024,
      "step": 1365
    },
    {
      "epoch": 0.062020416940175196,
      "grad_norm": 1.252446174621582,
      "learning_rate": 4.985562957546089e-05,
      "loss": 0.2348,
      "num_input_tokens_seen": 2204736,
      "step": 1370
    },
    {
      "epoch": 0.06224676882681817,
      "grad_norm": 0.758464515209198,
      "learning_rate": 4.9854574106649686e-05,
      "loss": 0.2398,
      "num_input_tokens_seen": 2212672,
      "step": 1375
    },
    {
      "epoch": 0.062473120713461144,
      "grad_norm": 0.4799676835536957,
      "learning_rate": 4.985351480495081e-05,
      "loss": 0.1913,
      "num_input_tokens_seen": 2220864,
      "step": 1380
    },
    {
      "epoch": 0.06269947260010413,
      "grad_norm": 0.43995410203933716,
      "learning_rate": 4.985245167052762e-05,
      "loss": 0.2024,
      "num_input_tokens_seen": 2228672,
      "step": 1385
    },
    {
      "epoch": 0.06292582448674709,
      "grad_norm": 0.7712388038635254,
      "learning_rate": 4.9851384703544066e-05,
      "loss": 0.2507,
      "num_input_tokens_seen": 2236448,
      "step": 1390
    },
    {
      "epoch": 0.06315217637339007,
      "grad_norm": 1.131034255027771,
      "learning_rate": 4.985031390416469e-05,
      "loss": 0.2358,
      "num_input_tokens_seen": 2243808,
      "step": 1395
    },
    {
      "epoch": 0.06337852826003305,
      "grad_norm": 0.7177644968032837,
      "learning_rate": 4.984923927255461e-05,
      "loss": 0.2108,
      "num_input_tokens_seen": 2252192,
      "step": 1400
    },
    {
      "epoch": 0.06337852826003305,
      "eval_loss": 0.22518377006053925,
      "eval_runtime": 403.6523,
      "eval_samples_per_second": 97.289,
      "eval_steps_per_second": 24.323,
      "num_input_tokens_seen": 2252192,
      "step": 1400
    },
    {
      "epoch": 0.06360488014667602,
      "grad_norm": 0.6828849911689758,
      "learning_rate": 4.984816080887958e-05,
      "loss": 0.256,
      "num_input_tokens_seen": 2261312,
      "step": 1405
    },
    {
      "epoch": 0.063831232033319,
      "grad_norm": 1.2374919652938843,
      "learning_rate": 4.9847078513305875e-05,
      "loss": 0.2249,
      "num_input_tokens_seen": 2269664,
      "step": 1410
    },
    {
      "epoch": 0.06405758391996197,
      "grad_norm": 0.8962686061859131,
      "learning_rate": 4.984599238600043e-05,
      "loss": 0.1904,
      "num_input_tokens_seen": 2278496,
      "step": 1415
    },
    {
      "epoch": 0.06428393580660495,
      "grad_norm": 0.7953397631645203,
      "learning_rate": 4.9844902427130716e-05,
      "loss": 0.2047,
      "num_input_tokens_seen": 2286368,
      "step": 1420
    },
    {
      "epoch": 0.06451028769324792,
      "grad_norm": 0.3667892515659332,
      "learning_rate": 4.984380863686482e-05,
      "loss": 0.2101,
      "num_input_tokens_seen": 2294624,
      "step": 1425
    },
    {
      "epoch": 0.0647366395798909,
      "grad_norm": 0.7745243310928345,
      "learning_rate": 4.984271101537143e-05,
      "loss": 0.2241,
      "num_input_tokens_seen": 2302720,
      "step": 1430
    },
    {
      "epoch": 0.06496299146653388,
      "grad_norm": 0.5526834726333618,
      "learning_rate": 4.9841609562819816e-05,
      "loss": 0.2397,
      "num_input_tokens_seen": 2310560,
      "step": 1435
    },
    {
      "epoch": 0.06518934335317685,
      "grad_norm": 1.1681382656097412,
      "learning_rate": 4.984050427937983e-05,
      "loss": 0.2503,
      "num_input_tokens_seen": 2319264,
      "step": 1440
    },
    {
      "epoch": 0.06541569523981983,
      "grad_norm": 0.9259237051010132,
      "learning_rate": 4.983939516522191e-05,
      "loss": 0.2147,
      "num_input_tokens_seen": 2327776,
      "step": 1445
    },
    {
      "epoch": 0.0656420471264628,
      "grad_norm": 0.4924405813217163,
      "learning_rate": 4.983828222051711e-05,
      "loss": 0.2481,
      "num_input_tokens_seen": 2335616,
      "step": 1450
    },
    {
      "epoch": 0.06586839901310577,
      "grad_norm": 0.7925808429718018,
      "learning_rate": 4.983716544543705e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 2343296,
      "step": 1455
    },
    {
      "epoch": 0.06609475089974876,
      "grad_norm": 0.49833008646965027,
      "learning_rate": 4.983604484015395e-05,
      "loss": 0.2632,
      "num_input_tokens_seen": 2352160,
      "step": 1460
    },
    {
      "epoch": 0.06632110278639172,
      "grad_norm": 0.6368857622146606,
      "learning_rate": 4.983492040484064e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 2360320,
      "step": 1465
    },
    {
      "epoch": 0.0665474546730347,
      "grad_norm": 0.9379906058311462,
      "learning_rate": 4.98337921396705e-05,
      "loss": 0.2287,
      "num_input_tokens_seen": 2367872,
      "step": 1470
    },
    {
      "epoch": 0.06677380655967767,
      "grad_norm": 0.6351608633995056,
      "learning_rate": 4.983266004481753e-05,
      "loss": 0.2515,
      "num_input_tokens_seen": 2376128,
      "step": 1475
    },
    {
      "epoch": 0.06700015844632065,
      "grad_norm": 0.45844557881355286,
      "learning_rate": 4.9831524120456316e-05,
      "loss": 0.2278,
      "num_input_tokens_seen": 2383872,
      "step": 1480
    },
    {
      "epoch": 0.06722651033296363,
      "grad_norm": 0.4570528566837311,
      "learning_rate": 4.9830384366762026e-05,
      "loss": 0.2064,
      "num_input_tokens_seen": 2392096,
      "step": 1485
    },
    {
      "epoch": 0.0674528622196066,
      "grad_norm": 0.8591161370277405,
      "learning_rate": 4.9829240783910436e-05,
      "loss": 0.2201,
      "num_input_tokens_seen": 2400256,
      "step": 1490
    },
    {
      "epoch": 0.06767921410624958,
      "grad_norm": 0.5506670475006104,
      "learning_rate": 4.982809337207789e-05,
      "loss": 0.2145,
      "num_input_tokens_seen": 2409216,
      "step": 1495
    },
    {
      "epoch": 0.06790556599289255,
      "grad_norm": 1.0913692712783813,
      "learning_rate": 4.9826942131441337e-05,
      "loss": 0.2493,
      "num_input_tokens_seen": 2417216,
      "step": 1500
    },
    {
      "epoch": 0.06813191787953553,
      "grad_norm": 0.44257116317749023,
      "learning_rate": 4.9825787062178315e-05,
      "loss": 0.2312,
      "num_input_tokens_seen": 2424992,
      "step": 1505
    },
    {
      "epoch": 0.0683582697661785,
      "grad_norm": 0.8103067874908447,
      "learning_rate": 4.9824628164466945e-05,
      "loss": 0.197,
      "num_input_tokens_seen": 2433888,
      "step": 1510
    },
    {
      "epoch": 0.06858462165282148,
      "grad_norm": 0.39938050508499146,
      "learning_rate": 4.982346543848595e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 2441888,
      "step": 1515
    },
    {
      "epoch": 0.06881097353946446,
      "grad_norm": 0.6128575801849365,
      "learning_rate": 4.9822298884414626e-05,
      "loss": 0.2179,
      "num_input_tokens_seen": 2449952,
      "step": 1520
    },
    {
      "epoch": 0.06903732542610742,
      "grad_norm": 0.8826938271522522,
      "learning_rate": 4.982112850243288e-05,
      "loss": 0.2358,
      "num_input_tokens_seen": 2457664,
      "step": 1525
    },
    {
      "epoch": 0.0692636773127504,
      "grad_norm": 0.9742387533187866,
      "learning_rate": 4.98199542927212e-05,
      "loss": 0.2256,
      "num_input_tokens_seen": 2465664,
      "step": 1530
    },
    {
      "epoch": 0.06949002919939337,
      "grad_norm": 0.6900627017021179,
      "learning_rate": 4.981877625546066e-05,
      "loss": 0.2165,
      "num_input_tokens_seen": 2473728,
      "step": 1535
    },
    {
      "epoch": 0.06971638108603635,
      "grad_norm": 0.5702961087226868,
      "learning_rate": 4.981759439083293e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 2481760,
      "step": 1540
    },
    {
      "epoch": 0.06994273297267933,
      "grad_norm": 0.6589258909225464,
      "learning_rate": 4.981640869902027e-05,
      "loss": 0.236,
      "num_input_tokens_seen": 2489696,
      "step": 1545
    },
    {
      "epoch": 0.0701690848593223,
      "grad_norm": 0.49065738916397095,
      "learning_rate": 4.9815219180205517e-05,
      "loss": 0.2215,
      "num_input_tokens_seen": 2497632,
      "step": 1550
    },
    {
      "epoch": 0.07039543674596528,
      "grad_norm": 0.7560986876487732,
      "learning_rate": 4.9814025834572126e-05,
      "loss": 0.2188,
      "num_input_tokens_seen": 2506048,
      "step": 1555
    },
    {
      "epoch": 0.07062178863260825,
      "grad_norm": 0.8792093396186829,
      "learning_rate": 4.981282866230411e-05,
      "loss": 0.2124,
      "num_input_tokens_seen": 2514176,
      "step": 1560
    },
    {
      "epoch": 0.07084814051925123,
      "grad_norm": 0.8318350911140442,
      "learning_rate": 4.981162766358611e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 2522208,
      "step": 1565
    },
    {
      "epoch": 0.0710744924058942,
      "grad_norm": 0.6064673066139221,
      "learning_rate": 4.9810422838603316e-05,
      "loss": 0.1919,
      "num_input_tokens_seen": 2530400,
      "step": 1570
    },
    {
      "epoch": 0.07130084429253718,
      "grad_norm": 1.0003958940505981,
      "learning_rate": 4.9809214187541533e-05,
      "loss": 0.2268,
      "num_input_tokens_seen": 2538336,
      "step": 1575
    },
    {
      "epoch": 0.07152719617918016,
      "grad_norm": 0.5913577079772949,
      "learning_rate": 4.980800171058715e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 2546048,
      "step": 1580
    },
    {
      "epoch": 0.07175354806582313,
      "grad_norm": 0.8752754926681519,
      "learning_rate": 4.980678540792715e-05,
      "loss": 0.2751,
      "num_input_tokens_seen": 2554016,
      "step": 1585
    },
    {
      "epoch": 0.0719798999524661,
      "grad_norm": 0.3361930549144745,
      "learning_rate": 4.980556527974909e-05,
      "loss": 0.2,
      "num_input_tokens_seen": 2562624,
      "step": 1590
    },
    {
      "epoch": 0.07220625183910907,
      "grad_norm": 0.6234817504882812,
      "learning_rate": 4.980434132624114e-05,
      "loss": 0.213,
      "num_input_tokens_seen": 2571264,
      "step": 1595
    },
    {
      "epoch": 0.07243260372575205,
      "grad_norm": 0.3948309123516083,
      "learning_rate": 4.980311354759205e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 2579136,
      "step": 1600
    },
    {
      "epoch": 0.07243260372575205,
      "eval_loss": 0.21877866983413696,
      "eval_runtime": 404.6438,
      "eval_samples_per_second": 97.051,
      "eval_steps_per_second": 24.263,
      "num_input_tokens_seen": 2579136,
      "step": 1600
    },
    {
      "epoch": 0.07265895561239503,
      "grad_norm": 0.5598514080047607,
      "learning_rate": 4.980188194399116e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 2587232,
      "step": 1605
    },
    {
      "epoch": 0.072885307499038,
      "grad_norm": 1.1014515161514282,
      "learning_rate": 4.9800646515628384e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 2595104,
      "step": 1610
    },
    {
      "epoch": 0.07311165938568098,
      "grad_norm": 0.7707337141036987,
      "learning_rate": 4.979940726269426e-05,
      "loss": 0.1972,
      "num_input_tokens_seen": 2602880,
      "step": 1615
    },
    {
      "epoch": 0.07333801127232395,
      "grad_norm": 0.4317009747028351,
      "learning_rate": 4.979816418537988e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 2611168,
      "step": 1620
    },
    {
      "epoch": 0.07356436315896693,
      "grad_norm": 0.8107607364654541,
      "learning_rate": 4.979691728387696e-05,
      "loss": 0.233,
      "num_input_tokens_seen": 2619168,
      "step": 1625
    },
    {
      "epoch": 0.07379071504560991,
      "grad_norm": 0.8154449462890625,
      "learning_rate": 4.979566655837776e-05,
      "loss": 0.2329,
      "num_input_tokens_seen": 2627808,
      "step": 1630
    },
    {
      "epoch": 0.07401706693225288,
      "grad_norm": 0.7444947957992554,
      "learning_rate": 4.9794412009075184e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 2635712,
      "step": 1635
    },
    {
      "epoch": 0.07424341881889586,
      "grad_norm": 0.33738335967063904,
      "learning_rate": 4.979315363616269e-05,
      "loss": 0.1857,
      "num_input_tokens_seen": 2643392,
      "step": 1640
    },
    {
      "epoch": 0.07446977070553883,
      "grad_norm": 0.9527921676635742,
      "learning_rate": 4.979189143983434e-05,
      "loss": 0.2063,
      "num_input_tokens_seen": 2651680,
      "step": 1645
    },
    {
      "epoch": 0.07469612259218181,
      "grad_norm": 0.6038247346878052,
      "learning_rate": 4.979062542028478e-05,
      "loss": 0.1983,
      "num_input_tokens_seen": 2659424,
      "step": 1650
    },
    {
      "epoch": 0.07492247447882477,
      "grad_norm": 0.4725152254104614,
      "learning_rate": 4.978935557770923e-05,
      "loss": 0.2418,
      "num_input_tokens_seen": 2667552,
      "step": 1655
    },
    {
      "epoch": 0.07514882636546776,
      "grad_norm": 0.819770097732544,
      "learning_rate": 4.978808191230353e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 2675424,
      "step": 1660
    },
    {
      "epoch": 0.07537517825211074,
      "grad_norm": 0.9013901948928833,
      "learning_rate": 4.9786804424264085e-05,
      "loss": 0.2227,
      "num_input_tokens_seen": 2683392,
      "step": 1665
    },
    {
      "epoch": 0.0756015301387537,
      "grad_norm": 0.40786877274513245,
      "learning_rate": 4.978552311378792e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 2691392,
      "step": 1670
    },
    {
      "epoch": 0.07582788202539668,
      "grad_norm": 0.7786810994148254,
      "learning_rate": 4.978423798107261e-05,
      "loss": 0.1945,
      "num_input_tokens_seen": 2699328,
      "step": 1675
    },
    {
      "epoch": 0.07605423391203965,
      "grad_norm": 1.1664087772369385,
      "learning_rate": 4.978294902631635e-05,
      "loss": 0.2064,
      "num_input_tokens_seen": 2707264,
      "step": 1680
    },
    {
      "epoch": 0.07628058579868263,
      "grad_norm": 0.5200538635253906,
      "learning_rate": 4.9781656249717914e-05,
      "loss": 0.1956,
      "num_input_tokens_seen": 2715520,
      "step": 1685
    },
    {
      "epoch": 0.07650693768532561,
      "grad_norm": 0.5482015013694763,
      "learning_rate": 4.9780359651476645e-05,
      "loss": 0.2389,
      "num_input_tokens_seen": 2723552,
      "step": 1690
    },
    {
      "epoch": 0.07673328957196858,
      "grad_norm": 0.2946873903274536,
      "learning_rate": 4.977905923179251e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 2731456,
      "step": 1695
    },
    {
      "epoch": 0.07695964145861156,
      "grad_norm": 0.3680018484592438,
      "learning_rate": 4.977775499086606e-05,
      "loss": 0.2069,
      "num_input_tokens_seen": 2739936,
      "step": 1700
    },
    {
      "epoch": 0.07718599334525453,
      "grad_norm": 0.6447925567626953,
      "learning_rate": 4.97764469288984e-05,
      "loss": 0.223,
      "num_input_tokens_seen": 2747904,
      "step": 1705
    },
    {
      "epoch": 0.07741234523189751,
      "grad_norm": 0.4183397591114044,
      "learning_rate": 4.977513504609127e-05,
      "loss": 0.2272,
      "num_input_tokens_seen": 2756256,
      "step": 1710
    },
    {
      "epoch": 0.07763869711854049,
      "grad_norm": 0.3283967673778534,
      "learning_rate": 4.9773819342646965e-05,
      "loss": 0.2379,
      "num_input_tokens_seen": 2764064,
      "step": 1715
    },
    {
      "epoch": 0.07786504900518346,
      "grad_norm": 0.3535142242908478,
      "learning_rate": 4.97724998187684e-05,
      "loss": 0.2226,
      "num_input_tokens_seen": 2772640,
      "step": 1720
    },
    {
      "epoch": 0.07809140089182644,
      "grad_norm": 0.49419158697128296,
      "learning_rate": 4.9771176474659045e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 2780288,
      "step": 1725
    },
    {
      "epoch": 0.0783177527784694,
      "grad_norm": 0.6159543991088867,
      "learning_rate": 4.976984931052299e-05,
      "loss": 0.2294,
      "num_input_tokens_seen": 2788576,
      "step": 1730
    },
    {
      "epoch": 0.07854410466511239,
      "grad_norm": 0.4340738356113434,
      "learning_rate": 4.976851832656489e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 2796544,
      "step": 1735
    },
    {
      "epoch": 0.07877045655175535,
      "grad_norm": 0.7000336647033691,
      "learning_rate": 4.9767183522990004e-05,
      "loss": 0.1984,
      "num_input_tokens_seen": 2804320,
      "step": 1740
    },
    {
      "epoch": 0.07899680843839833,
      "grad_norm": 0.7558102607727051,
      "learning_rate": 4.9765844900004176e-05,
      "loss": 0.2606,
      "num_input_tokens_seen": 2812160,
      "step": 1745
    },
    {
      "epoch": 0.07922316032504131,
      "grad_norm": 0.36175334453582764,
      "learning_rate": 4.9764502457813834e-05,
      "loss": 0.2202,
      "num_input_tokens_seen": 2819712,
      "step": 1750
    },
    {
      "epoch": 0.07944951221168428,
      "grad_norm": 0.6707577109336853,
      "learning_rate": 4.9763156196626005e-05,
      "loss": 0.2075,
      "num_input_tokens_seen": 2827392,
      "step": 1755
    },
    {
      "epoch": 0.07967586409832726,
      "grad_norm": 0.4860638380050659,
      "learning_rate": 4.97618061166483e-05,
      "loss": 0.2216,
      "num_input_tokens_seen": 2835520,
      "step": 1760
    },
    {
      "epoch": 0.07990221598497023,
      "grad_norm": 0.5759274959564209,
      "learning_rate": 4.9760452218088915e-05,
      "loss": 0.2368,
      "num_input_tokens_seen": 2843008,
      "step": 1765
    },
    {
      "epoch": 0.08012856787161321,
      "grad_norm": 0.40114203095436096,
      "learning_rate": 4.975909450115663e-05,
      "loss": 0.2249,
      "num_input_tokens_seen": 2850752,
      "step": 1770
    },
    {
      "epoch": 0.08035491975825619,
      "grad_norm": 0.7788422703742981,
      "learning_rate": 4.975773296606084e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 2859264,
      "step": 1775
    },
    {
      "epoch": 0.08058127164489916,
      "grad_norm": 0.5490627288818359,
      "learning_rate": 4.97563676130115e-05,
      "loss": 0.2213,
      "num_input_tokens_seen": 2867872,
      "step": 1780
    },
    {
      "epoch": 0.08080762353154214,
      "grad_norm": 0.8881388902664185,
      "learning_rate": 4.9754998442219166e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 2875168,
      "step": 1785
    },
    {
      "epoch": 0.0810339754181851,
      "grad_norm": 0.8983529806137085,
      "learning_rate": 4.9753625453894984e-05,
      "loss": 0.2015,
      "num_input_tokens_seen": 2882688,
      "step": 1790
    },
    {
      "epoch": 0.08126032730482809,
      "grad_norm": 0.5461991429328918,
      "learning_rate": 4.975224864825068e-05,
      "loss": 0.2194,
      "num_input_tokens_seen": 2890816,
      "step": 1795
    },
    {
      "epoch": 0.08148667919147107,
      "grad_norm": 0.5056917071342468,
      "learning_rate": 4.9750868025498576e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 2898880,
      "step": 1800
    },
    {
      "epoch": 0.08148667919147107,
      "eval_loss": 0.21393464505672455,
      "eval_runtime": 404.7152,
      "eval_samples_per_second": 97.034,
      "eval_steps_per_second": 24.259,
      "num_input_tokens_seen": 2898880,
      "step": 1800
    },
    {
      "epoch": 0.08171303107811403,
      "grad_norm": 0.41253232955932617,
      "learning_rate": 4.974948358585158e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 2907200,
      "step": 1805
    },
    {
      "epoch": 0.08193938296475702,
      "grad_norm": 0.4507910907268524,
      "learning_rate": 4.9748095329523205e-05,
      "loss": 0.2164,
      "num_input_tokens_seen": 2915168,
      "step": 1810
    },
    {
      "epoch": 0.08216573485139998,
      "grad_norm": 0.6284390687942505,
      "learning_rate": 4.974670325672752e-05,
      "loss": 0.188,
      "num_input_tokens_seen": 2922848,
      "step": 1815
    },
    {
      "epoch": 0.08239208673804296,
      "grad_norm": 0.7267695069313049,
      "learning_rate": 4.974530736767921e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 2931104,
      "step": 1820
    },
    {
      "epoch": 0.08261843862468593,
      "grad_norm": 0.6925939321517944,
      "learning_rate": 4.9743907662593524e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 2939104,
      "step": 1825
    },
    {
      "epoch": 0.08284479051132891,
      "grad_norm": 1.1118348836898804,
      "learning_rate": 4.974250414168633e-05,
      "loss": 0.1924,
      "num_input_tokens_seen": 2947616,
      "step": 1830
    },
    {
      "epoch": 0.08307114239797189,
      "grad_norm": 0.632513701915741,
      "learning_rate": 4.974109680517407e-05,
      "loss": 0.2236,
      "num_input_tokens_seen": 2955872,
      "step": 1835
    },
    {
      "epoch": 0.08329749428461486,
      "grad_norm": 0.4011169970035553,
      "learning_rate": 4.973968565327376e-05,
      "loss": 0.2377,
      "num_input_tokens_seen": 2964640,
      "step": 1840
    },
    {
      "epoch": 0.08352384617125784,
      "grad_norm": 0.40770092606544495,
      "learning_rate": 4.973827068620303e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 2972768,
      "step": 1845
    },
    {
      "epoch": 0.08375019805790081,
      "grad_norm": 1.469234824180603,
      "learning_rate": 4.973685190418008e-05,
      "loss": 0.267,
      "num_input_tokens_seen": 2980896,
      "step": 1850
    },
    {
      "epoch": 0.08397654994454379,
      "grad_norm": 0.32841256260871887,
      "learning_rate": 4.97354293074237e-05,
      "loss": 0.2561,
      "num_input_tokens_seen": 2988864,
      "step": 1855
    },
    {
      "epoch": 0.08420290183118677,
      "grad_norm": 1.103104829788208,
      "learning_rate": 4.9734002896153276e-05,
      "loss": 0.2435,
      "num_input_tokens_seen": 2996640,
      "step": 1860
    },
    {
      "epoch": 0.08442925371782974,
      "grad_norm": 0.43709489703178406,
      "learning_rate": 4.973257267058877e-05,
      "loss": 0.2113,
      "num_input_tokens_seen": 3005440,
      "step": 1865
    },
    {
      "epoch": 0.08465560560447272,
      "grad_norm": 0.3825618326663971,
      "learning_rate": 4.973113863095076e-05,
      "loss": 0.2575,
      "num_input_tokens_seen": 3013184,
      "step": 1870
    },
    {
      "epoch": 0.08488195749111568,
      "grad_norm": 0.3902733623981476,
      "learning_rate": 4.9729700777460384e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 3020640,
      "step": 1875
    },
    {
      "epoch": 0.08510830937775866,
      "grad_norm": 0.47591447830200195,
      "learning_rate": 4.972825911033937e-05,
      "loss": 0.2464,
      "num_input_tokens_seen": 3028416,
      "step": 1880
    },
    {
      "epoch": 0.08533466126440165,
      "grad_norm": 0.7358670234680176,
      "learning_rate": 4.9726813629810056e-05,
      "loss": 0.2349,
      "num_input_tokens_seen": 3037120,
      "step": 1885
    },
    {
      "epoch": 0.08556101315104461,
      "grad_norm": 0.5646424889564514,
      "learning_rate": 4.9725364336095326e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 3044512,
      "step": 1890
    },
    {
      "epoch": 0.0857873650376876,
      "grad_norm": 0.4208298325538635,
      "learning_rate": 4.972391122941871e-05,
      "loss": 0.2208,
      "num_input_tokens_seen": 3052800,
      "step": 1895
    },
    {
      "epoch": 0.08601371692433056,
      "grad_norm": 0.34525159001350403,
      "learning_rate": 4.972245431000428e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 3061280,
      "step": 1900
    },
    {
      "epoch": 0.08624006881097354,
      "grad_norm": 0.3941461443901062,
      "learning_rate": 4.972099357807671e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 3068928,
      "step": 1905
    },
    {
      "epoch": 0.08646642069761651,
      "grad_norm": 0.49912598729133606,
      "learning_rate": 4.971952903386127e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 3076640,
      "step": 1910
    },
    {
      "epoch": 0.08669277258425949,
      "grad_norm": 0.4011307656764984,
      "learning_rate": 4.971806067758381e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 3084832,
      "step": 1915
    },
    {
      "epoch": 0.08691912447090247,
      "grad_norm": 0.6464841961860657,
      "learning_rate": 4.971658850947076e-05,
      "loss": 0.2053,
      "num_input_tokens_seen": 3092672,
      "step": 1920
    },
    {
      "epoch": 0.08714547635754544,
      "grad_norm": 0.70502769947052,
      "learning_rate": 4.9715112529749165e-05,
      "loss": 0.2336,
      "num_input_tokens_seen": 3100384,
      "step": 1925
    },
    {
      "epoch": 0.08737182824418842,
      "grad_norm": 1.1347774267196655,
      "learning_rate": 4.9713632738646624e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 3109024,
      "step": 1930
    },
    {
      "epoch": 0.08759818013083139,
      "grad_norm": 0.4608071446418762,
      "learning_rate": 4.971214913639134e-05,
      "loss": 0.208,
      "num_input_tokens_seen": 3116736,
      "step": 1935
    },
    {
      "epoch": 0.08782453201747437,
      "grad_norm": 0.8622046113014221,
      "learning_rate": 4.9710661723212104e-05,
      "loss": 0.1943,
      "num_input_tokens_seen": 3124768,
      "step": 1940
    },
    {
      "epoch": 0.08805088390411735,
      "grad_norm": 0.4367600083351135,
      "learning_rate": 4.9709170499338295e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 3132576,
      "step": 1945
    },
    {
      "epoch": 0.08827723579076031,
      "grad_norm": 0.3858290910720825,
      "learning_rate": 4.9707675464999895e-05,
      "loss": 0.2193,
      "num_input_tokens_seen": 3140384,
      "step": 1950
    },
    {
      "epoch": 0.0885035876774033,
      "grad_norm": 1.2259188890457153,
      "learning_rate": 4.970617662042743e-05,
      "loss": 0.216,
      "num_input_tokens_seen": 3148768,
      "step": 1955
    },
    {
      "epoch": 0.08872993956404626,
      "grad_norm": 0.5257636308670044,
      "learning_rate": 4.970467396585206e-05,
      "loss": 0.1962,
      "num_input_tokens_seen": 3156992,
      "step": 1960
    },
    {
      "epoch": 0.08895629145068924,
      "grad_norm": 0.5516003370285034,
      "learning_rate": 4.97031675015055e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 3165376,
      "step": 1965
    },
    {
      "epoch": 0.08918264333733222,
      "grad_norm": 0.48239296674728394,
      "learning_rate": 4.9701657227620075e-05,
      "loss": 0.2135,
      "num_input_tokens_seen": 3173856,
      "step": 1970
    },
    {
      "epoch": 0.08940899522397519,
      "grad_norm": 0.4637291431427002,
      "learning_rate": 4.9700143144428685e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 3181760,
      "step": 1975
    },
    {
      "epoch": 0.08963534711061817,
      "grad_norm": 0.7839916944503784,
      "learning_rate": 4.969862525216482e-05,
      "loss": 0.2235,
      "num_input_tokens_seen": 3190048,
      "step": 1980
    },
    {
      "epoch": 0.08986169899726114,
      "grad_norm": 0.6278282403945923,
      "learning_rate": 4.9697103551062556e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 3197728,
      "step": 1985
    },
    {
      "epoch": 0.09008805088390412,
      "grad_norm": 0.8005794286727905,
      "learning_rate": 4.9695578041356565e-05,
      "loss": 0.2013,
      "num_input_tokens_seen": 3205312,
      "step": 1990
    },
    {
      "epoch": 0.09031440277054709,
      "grad_norm": 0.9282761812210083,
      "learning_rate": 4.969404872328209e-05,
      "loss": 0.2238,
      "num_input_tokens_seen": 3213408,
      "step": 1995
    },
    {
      "epoch": 0.09054075465719007,
      "grad_norm": 0.813024640083313,
      "learning_rate": 4.969251559707498e-05,
      "loss": 0.2349,
      "num_input_tokens_seen": 3221312,
      "step": 2000
    },
    {
      "epoch": 0.09054075465719007,
      "eval_loss": 0.2094564437866211,
      "eval_runtime": 404.064,
      "eval_samples_per_second": 97.19,
      "eval_steps_per_second": 24.298,
      "num_input_tokens_seen": 3221312,
      "step": 2000
    },
    {
      "epoch": 0.09076710654383305,
      "grad_norm": 0.6020380854606628,
      "learning_rate": 4.9690978662971674e-05,
      "loss": 0.2097,
      "num_input_tokens_seen": 3228608,
      "step": 2005
    },
    {
      "epoch": 0.09099345843047602,
      "grad_norm": 0.3669753670692444,
      "learning_rate": 4.968943792120916e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 3236544,
      "step": 2010
    },
    {
      "epoch": 0.091219810317119,
      "grad_norm": 0.33654776215553284,
      "learning_rate": 4.9687893372025046e-05,
      "loss": 0.1988,
      "num_input_tokens_seen": 3245120,
      "step": 2015
    },
    {
      "epoch": 0.09144616220376196,
      "grad_norm": 1.2802098989486694,
      "learning_rate": 4.9686345015657535e-05,
      "loss": 0.2206,
      "num_input_tokens_seen": 3253024,
      "step": 2020
    },
    {
      "epoch": 0.09167251409040494,
      "grad_norm": 0.3832942843437195,
      "learning_rate": 4.968479285234538e-05,
      "loss": 0.2129,
      "num_input_tokens_seen": 3261280,
      "step": 2025
    },
    {
      "epoch": 0.09189886597704792,
      "grad_norm": 0.35121458768844604,
      "learning_rate": 4.9683236882327974e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 3269184,
      "step": 2030
    },
    {
      "epoch": 0.09212521786369089,
      "grad_norm": 0.43943071365356445,
      "learning_rate": 4.968167710584526e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 3277376,
      "step": 2035
    },
    {
      "epoch": 0.09235156975033387,
      "grad_norm": 1.0928611755371094,
      "learning_rate": 4.968011352313775e-05,
      "loss": 0.2132,
      "num_input_tokens_seen": 3285376,
      "step": 2040
    },
    {
      "epoch": 0.09257792163697684,
      "grad_norm": 0.46686413884162903,
      "learning_rate": 4.967854613444659e-05,
      "loss": 0.2015,
      "num_input_tokens_seen": 3294016,
      "step": 2045
    },
    {
      "epoch": 0.09280427352361982,
      "grad_norm": 0.5331802368164062,
      "learning_rate": 4.967697494001349e-05,
      "loss": 0.212,
      "num_input_tokens_seen": 3301760,
      "step": 2050
    },
    {
      "epoch": 0.09303062541026279,
      "grad_norm": 1.0260396003723145,
      "learning_rate": 4.9675399940080736e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 3309856,
      "step": 2055
    },
    {
      "epoch": 0.09325697729690577,
      "grad_norm": 0.49788057804107666,
      "learning_rate": 4.9673821134891226e-05,
      "loss": 0.2016,
      "num_input_tokens_seen": 3317984,
      "step": 2060
    },
    {
      "epoch": 0.09348332918354875,
      "grad_norm": 0.90938800573349,
      "learning_rate": 4.967223852468842e-05,
      "loss": 0.2012,
      "num_input_tokens_seen": 3326176,
      "step": 2065
    },
    {
      "epoch": 0.09370968107019172,
      "grad_norm": 0.7520653009414673,
      "learning_rate": 4.967065210971639e-05,
      "loss": 0.2037,
      "num_input_tokens_seen": 3334080,
      "step": 2070
    },
    {
      "epoch": 0.0939360329568347,
      "grad_norm": 0.4723113775253296,
      "learning_rate": 4.966906189021977e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 3342112,
      "step": 2075
    },
    {
      "epoch": 0.09416238484347766,
      "grad_norm": 0.2697027325630188,
      "learning_rate": 4.966746786644379e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 3349760,
      "step": 2080
    },
    {
      "epoch": 0.09438873673012065,
      "grad_norm": 0.7084248661994934,
      "learning_rate": 4.966587003863429e-05,
      "loss": 0.2236,
      "num_input_tokens_seen": 3358080,
      "step": 2085
    },
    {
      "epoch": 0.09461508861676363,
      "grad_norm": 0.6124038696289062,
      "learning_rate": 4.966426840703765e-05,
      "loss": 0.1968,
      "num_input_tokens_seen": 3365632,
      "step": 2090
    },
    {
      "epoch": 0.0948414405034066,
      "grad_norm": 0.7037796378135681,
      "learning_rate": 4.9662662971900875e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 3373280,
      "step": 2095
    },
    {
      "epoch": 0.09506779239004957,
      "grad_norm": 0.42623037099838257,
      "learning_rate": 4.9661053733471534e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 3381344,
      "step": 2100
    },
    {
      "epoch": 0.09529414427669254,
      "grad_norm": 0.45579466223716736,
      "learning_rate": 4.965944069199781e-05,
      "loss": 0.2437,
      "num_input_tokens_seen": 3389056,
      "step": 2105
    },
    {
      "epoch": 0.09552049616333552,
      "grad_norm": 0.5080585479736328,
      "learning_rate": 4.965782384772842e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 3396960,
      "step": 2110
    },
    {
      "epoch": 0.0957468480499785,
      "grad_norm": 0.7721723914146423,
      "learning_rate": 4.9656203200912734e-05,
      "loss": 0.2343,
      "num_input_tokens_seen": 3405088,
      "step": 2115
    },
    {
      "epoch": 0.09597319993662147,
      "grad_norm": 0.5496774315834045,
      "learning_rate": 4.965457875180067e-05,
      "loss": 0.2044,
      "num_input_tokens_seen": 3412960,
      "step": 2120
    },
    {
      "epoch": 0.09619955182326445,
      "grad_norm": 0.6188136339187622,
      "learning_rate": 4.9652950500642724e-05,
      "loss": 0.2454,
      "num_input_tokens_seen": 3421600,
      "step": 2125
    },
    {
      "epoch": 0.09642590370990742,
      "grad_norm": 0.3012542426586151,
      "learning_rate": 4.965131844769001e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 3429280,
      "step": 2130
    },
    {
      "epoch": 0.0966522555965504,
      "grad_norm": 0.6210690140724182,
      "learning_rate": 4.96496825931942e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 3436960,
      "step": 2135
    },
    {
      "epoch": 0.09687860748319337,
      "grad_norm": 0.7577105164527893,
      "learning_rate": 4.9648042937407566e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 3446336,
      "step": 2140
    },
    {
      "epoch": 0.09710495936983635,
      "grad_norm": 1.0634410381317139,
      "learning_rate": 4.964639948058297e-05,
      "loss": 0.2436,
      "num_input_tokens_seen": 3454432,
      "step": 2145
    },
    {
      "epoch": 0.09733131125647933,
      "grad_norm": 0.5898555517196655,
      "learning_rate": 4.9644752222973846e-05,
      "loss": 0.1938,
      "num_input_tokens_seen": 3462560,
      "step": 2150
    },
    {
      "epoch": 0.0975576631431223,
      "grad_norm": 0.3136095106601715,
      "learning_rate": 4.964310116483422e-05,
      "loss": 0.2085,
      "num_input_tokens_seen": 3470432,
      "step": 2155
    },
    {
      "epoch": 0.09778401502976528,
      "grad_norm": 0.5219261646270752,
      "learning_rate": 4.964144630641872e-05,
      "loss": 0.201,
      "num_input_tokens_seen": 3478496,
      "step": 2160
    },
    {
      "epoch": 0.09801036691640824,
      "grad_norm": 0.7368583083152771,
      "learning_rate": 4.9639787647982525e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 3486176,
      "step": 2165
    },
    {
      "epoch": 0.09823671880305122,
      "grad_norm": 0.37081974744796753,
      "learning_rate": 4.963812518978143e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 3494112,
      "step": 2170
    },
    {
      "epoch": 0.0984630706896942,
      "grad_norm": 0.9466499090194702,
      "learning_rate": 4.963645893207182e-05,
      "loss": 0.2422,
      "num_input_tokens_seen": 3501920,
      "step": 2175
    },
    {
      "epoch": 0.09868942257633717,
      "grad_norm": 0.8841801881790161,
      "learning_rate": 4.963478887511063e-05,
      "loss": 0.2048,
      "num_input_tokens_seen": 3509728,
      "step": 2180
    },
    {
      "epoch": 0.09891577446298015,
      "grad_norm": 0.2508432865142822,
      "learning_rate": 4.963311501915542e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 3517696,
      "step": 2185
    },
    {
      "epoch": 0.09914212634962312,
      "grad_norm": 0.5913707613945007,
      "learning_rate": 4.963143736446432e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 3526592,
      "step": 2190
    },
    {
      "epoch": 0.0993684782362661,
      "grad_norm": 0.7151490449905396,
      "learning_rate": 4.962975591129603e-05,
      "loss": 0.2176,
      "num_input_tokens_seen": 3534816,
      "step": 2195
    },
    {
      "epoch": 0.09959483012290908,
      "grad_norm": 0.888482391834259,
      "learning_rate": 4.962807065990986e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 3542752,
      "step": 2200
    },
    {
      "epoch": 0.09959483012290908,
      "eval_loss": 0.20747481286525726,
      "eval_runtime": 404.9193,
      "eval_samples_per_second": 96.985,
      "eval_steps_per_second": 24.247,
      "num_input_tokens_seen": 3542752,
      "step": 2200
    },
    {
      "epoch": 0.09982118200955205,
      "grad_norm": 0.46328338980674744,
      "learning_rate": 4.9626381610565714e-05,
      "loss": 0.1949,
      "num_input_tokens_seen": 3550784,
      "step": 2205
    },
    {
      "epoch": 0.10004753389619503,
      "grad_norm": 0.5599423050880432,
      "learning_rate": 4.9624688763524043e-05,
      "loss": 0.2523,
      "num_input_tokens_seen": 3558880,
      "step": 2210
    },
    {
      "epoch": 0.100273885782838,
      "grad_norm": 0.6383845806121826,
      "learning_rate": 4.962299211904591e-05,
      "loss": 0.2242,
      "num_input_tokens_seen": 3567456,
      "step": 2215
    },
    {
      "epoch": 0.10050023766948098,
      "grad_norm": 0.7692732214927673,
      "learning_rate": 4.962129167739296e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 3575104,
      "step": 2220
    },
    {
      "epoch": 0.10072658955612394,
      "grad_norm": 0.7509031295776367,
      "learning_rate": 4.961958743882742e-05,
      "loss": 0.2636,
      "num_input_tokens_seen": 3583200,
      "step": 2225
    },
    {
      "epoch": 0.10095294144276692,
      "grad_norm": 0.39978498220443726,
      "learning_rate": 4.961787940361211e-05,
      "loss": 0.2159,
      "num_input_tokens_seen": 3591456,
      "step": 2230
    },
    {
      "epoch": 0.1011792933294099,
      "grad_norm": 1.4787871837615967,
      "learning_rate": 4.961616757201043e-05,
      "loss": 0.1812,
      "num_input_tokens_seen": 3599648,
      "step": 2235
    },
    {
      "epoch": 0.10140564521605287,
      "grad_norm": 0.4525032043457031,
      "learning_rate": 4.961445194428637e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 3607680,
      "step": 2240
    },
    {
      "epoch": 0.10163199710269585,
      "grad_norm": 0.5850090980529785,
      "learning_rate": 4.9612732520704486e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 3615552,
      "step": 2245
    },
    {
      "epoch": 0.10185834898933882,
      "grad_norm": 0.42262202501296997,
      "learning_rate": 4.961100930152994e-05,
      "loss": 0.2185,
      "num_input_tokens_seen": 3623296,
      "step": 2250
    },
    {
      "epoch": 0.1020847008759818,
      "grad_norm": 0.5803593993186951,
      "learning_rate": 4.960928228702849e-05,
      "loss": 0.2225,
      "num_input_tokens_seen": 3632160,
      "step": 2255
    },
    {
      "epoch": 0.10231105276262478,
      "grad_norm": 0.6251130700111389,
      "learning_rate": 4.960755147746645e-05,
      "loss": 0.2062,
      "num_input_tokens_seen": 3640064,
      "step": 2260
    },
    {
      "epoch": 0.10253740464926775,
      "grad_norm": 0.43355652689933777,
      "learning_rate": 4.9605816873110736e-05,
      "loss": 0.2036,
      "num_input_tokens_seen": 3647872,
      "step": 2265
    },
    {
      "epoch": 0.10276375653591073,
      "grad_norm": 0.600823700428009,
      "learning_rate": 4.960407847422883e-05,
      "loss": 0.2032,
      "num_input_tokens_seen": 3655840,
      "step": 2270
    },
    {
      "epoch": 0.1029901084225537,
      "grad_norm": 0.8933124542236328,
      "learning_rate": 4.960233628108885e-05,
      "loss": 0.1907,
      "num_input_tokens_seen": 3663936,
      "step": 2275
    },
    {
      "epoch": 0.10321646030919668,
      "grad_norm": 0.6298086643218994,
      "learning_rate": 4.960059029395942e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 3672128,
      "step": 2280
    },
    {
      "epoch": 0.10344281219583966,
      "grad_norm": 0.6022596955299377,
      "learning_rate": 4.959884051310983e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 3679936,
      "step": 2285
    },
    {
      "epoch": 0.10366916408248263,
      "grad_norm": 0.6035820245742798,
      "learning_rate": 4.959708693880991e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 3689376,
      "step": 2290
    },
    {
      "epoch": 0.1038955159691256,
      "grad_norm": 0.7120176553726196,
      "learning_rate": 4.9595329571330074e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 3697920,
      "step": 2295
    },
    {
      "epoch": 0.10412186785576857,
      "grad_norm": 0.6699591875076294,
      "learning_rate": 4.9593568410941326e-05,
      "loss": 0.2011,
      "num_input_tokens_seen": 3705888,
      "step": 2300
    },
    {
      "epoch": 0.10434821974241155,
      "grad_norm": 0.5837795734405518,
      "learning_rate": 4.959180345791528e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 3713664,
      "step": 2305
    },
    {
      "epoch": 0.10457457162905452,
      "grad_norm": 0.539732813835144,
      "learning_rate": 4.9590034712524086e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 3721568,
      "step": 2310
    },
    {
      "epoch": 0.1048009235156975,
      "grad_norm": 0.6811764240264893,
      "learning_rate": 4.958826217504053e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 3729312,
      "step": 2315
    },
    {
      "epoch": 0.10502727540234048,
      "grad_norm": 0.4933869242668152,
      "learning_rate": 4.958648584573795e-05,
      "loss": 0.2125,
      "num_input_tokens_seen": 3737568,
      "step": 2320
    },
    {
      "epoch": 0.10525362728898345,
      "grad_norm": 0.8062941431999207,
      "learning_rate": 4.958470572489028e-05,
      "loss": 0.221,
      "num_input_tokens_seen": 3745376,
      "step": 2325
    },
    {
      "epoch": 0.10547997917562643,
      "grad_norm": 0.36250364780426025,
      "learning_rate": 4.958292181277203e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 3752864,
      "step": 2330
    },
    {
      "epoch": 0.1057063310622694,
      "grad_norm": 0.6925029754638672,
      "learning_rate": 4.958113410965832e-05,
      "loss": 0.2359,
      "num_input_tokens_seen": 3760992,
      "step": 2335
    },
    {
      "epoch": 0.10593268294891238,
      "grad_norm": 1.0215387344360352,
      "learning_rate": 4.957934261582481e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 3768928,
      "step": 2340
    },
    {
      "epoch": 0.10615903483555536,
      "grad_norm": 0.5416283011436462,
      "learning_rate": 4.95775473315478e-05,
      "loss": 0.2148,
      "num_input_tokens_seen": 3777024,
      "step": 2345
    },
    {
      "epoch": 0.10638538672219833,
      "grad_norm": 0.3477165102958679,
      "learning_rate": 4.9575748257104124e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 3785600,
      "step": 2350
    },
    {
      "epoch": 0.10661173860884131,
      "grad_norm": 0.5056434273719788,
      "learning_rate": 4.9573945392771224e-05,
      "loss": 0.2463,
      "num_input_tokens_seen": 3793568,
      "step": 2355
    },
    {
      "epoch": 0.10683809049548428,
      "grad_norm": 0.45973727107048035,
      "learning_rate": 4.9572138738827134e-05,
      "loss": 0.1921,
      "num_input_tokens_seen": 3801472,
      "step": 2360
    },
    {
      "epoch": 0.10706444238212726,
      "grad_norm": 0.5103929042816162,
      "learning_rate": 4.957032829555046e-05,
      "loss": 0.1958,
      "num_input_tokens_seen": 3809792,
      "step": 2365
    },
    {
      "epoch": 0.10729079426877024,
      "grad_norm": 0.3898981213569641,
      "learning_rate": 4.956851406322039e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 3817760,
      "step": 2370
    },
    {
      "epoch": 0.1075171461554132,
      "grad_norm": 0.5204512476921082,
      "learning_rate": 4.9566696042116704e-05,
      "loss": 0.2212,
      "num_input_tokens_seen": 3825536,
      "step": 2375
    },
    {
      "epoch": 0.10774349804205618,
      "grad_norm": 0.42835313081741333,
      "learning_rate": 4.9564874232519766e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 3833504,
      "step": 2380
    },
    {
      "epoch": 0.10796984992869915,
      "grad_norm": 0.520008385181427,
      "learning_rate": 4.9563048634710516e-05,
      "loss": 0.2363,
      "num_input_tokens_seen": 3842080,
      "step": 2385
    },
    {
      "epoch": 0.10819620181534213,
      "grad_norm": 0.7546582221984863,
      "learning_rate": 4.956121924897049e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 3850208,
      "step": 2390
    },
    {
      "epoch": 0.1084225537019851,
      "grad_norm": 0.530505359172821,
      "learning_rate": 4.955938607558181e-05,
      "loss": 0.1821,
      "num_input_tokens_seen": 3858400,
      "step": 2395
    },
    {
      "epoch": 0.10864890558862808,
      "grad_norm": 0.5338477492332458,
      "learning_rate": 4.955754911482715e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 3866688,
      "step": 2400
    },
    {
      "epoch": 0.10864890558862808,
      "eval_loss": 0.2019547075033188,
      "eval_runtime": 404.7554,
      "eval_samples_per_second": 97.024,
      "eval_steps_per_second": 24.257,
      "num_input_tokens_seen": 3866688,
      "step": 2400
    },
    {
      "epoch": 0.10887525747527106,
      "grad_norm": 0.7265180349349976,
      "learning_rate": 4.9555708366989804e-05,
      "loss": 0.2069,
      "num_input_tokens_seen": 3875168,
      "step": 2405
    },
    {
      "epoch": 0.10910160936191403,
      "grad_norm": 0.6813026070594788,
      "learning_rate": 4.9553863832353655e-05,
      "loss": 0.2135,
      "num_input_tokens_seen": 3883232,
      "step": 2410
    },
    {
      "epoch": 0.10932796124855701,
      "grad_norm": 0.4180748164653778,
      "learning_rate": 4.955201551120313e-05,
      "loss": 0.2153,
      "num_input_tokens_seen": 3891008,
      "step": 2415
    },
    {
      "epoch": 0.10955431313519998,
      "grad_norm": 0.5851491093635559,
      "learning_rate": 4.955016340382328e-05,
      "loss": 0.1958,
      "num_input_tokens_seen": 3898752,
      "step": 2420
    },
    {
      "epoch": 0.10978066502184296,
      "grad_norm": 0.9094352722167969,
      "learning_rate": 4.954830751049972e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 3906272,
      "step": 2425
    },
    {
      "epoch": 0.11000701690848594,
      "grad_norm": 0.5470118522644043,
      "learning_rate": 4.954644783151864e-05,
      "loss": 0.1994,
      "num_input_tokens_seen": 3914112,
      "step": 2430
    },
    {
      "epoch": 0.1102333687951289,
      "grad_norm": 0.516863226890564,
      "learning_rate": 4.954458436716684e-05,
      "loss": 0.2253,
      "num_input_tokens_seen": 3922016,
      "step": 2435
    },
    {
      "epoch": 0.11045972068177189,
      "grad_norm": 1.0075509548187256,
      "learning_rate": 4.954271711773168e-05,
      "loss": 0.2084,
      "num_input_tokens_seen": 3930336,
      "step": 2440
    },
    {
      "epoch": 0.11068607256841485,
      "grad_norm": 1.1205600500106812,
      "learning_rate": 4.9540846083501115e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 3938144,
      "step": 2445
    },
    {
      "epoch": 0.11091242445505783,
      "grad_norm": 0.7228522896766663,
      "learning_rate": 4.953897126476369e-05,
      "loss": 0.1707,
      "num_input_tokens_seen": 3946688,
      "step": 2450
    },
    {
      "epoch": 0.1111387763417008,
      "grad_norm": 0.9257678389549255,
      "learning_rate": 4.9537092661808514e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 3954976,
      "step": 2455
    },
    {
      "epoch": 0.11136512822834378,
      "grad_norm": 0.5313231348991394,
      "learning_rate": 4.9535210274925306e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 3963232,
      "step": 2460
    },
    {
      "epoch": 0.11159148011498676,
      "grad_norm": 0.3189002275466919,
      "learning_rate": 4.953332410440435e-05,
      "loss": 0.2402,
      "num_input_tokens_seen": 3971264,
      "step": 2465
    },
    {
      "epoch": 0.11181783200162973,
      "grad_norm": 0.6780828833580017,
      "learning_rate": 4.9531434150536496e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 3979584,
      "step": 2470
    },
    {
      "epoch": 0.11204418388827271,
      "grad_norm": 0.47021451592445374,
      "learning_rate": 4.952954041361322e-05,
      "loss": 0.2247,
      "num_input_tokens_seen": 3987456,
      "step": 2475
    },
    {
      "epoch": 0.11227053577491568,
      "grad_norm": 1.1562378406524658,
      "learning_rate": 4.952764289392655e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 3995488,
      "step": 2480
    },
    {
      "epoch": 0.11249688766155866,
      "grad_norm": 0.8544884920120239,
      "learning_rate": 4.952574159176912e-05,
      "loss": 0.2283,
      "num_input_tokens_seen": 4003456,
      "step": 2485
    },
    {
      "epoch": 0.11272323954820164,
      "grad_norm": 0.36219364404678345,
      "learning_rate": 4.952383650743413e-05,
      "loss": 0.1967,
      "num_input_tokens_seen": 4012288,
      "step": 2490
    },
    {
      "epoch": 0.1129495914348446,
      "grad_norm": 0.32383614778518677,
      "learning_rate": 4.952192764121536e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 4020640,
      "step": 2495
    },
    {
      "epoch": 0.11317594332148759,
      "grad_norm": 0.9219053387641907,
      "learning_rate": 4.9520014993407185e-05,
      "loss": 0.224,
      "num_input_tokens_seen": 4028544,
      "step": 2500
    },
    {
      "epoch": 0.11340229520813055,
      "grad_norm": 0.32820701599121094,
      "learning_rate": 4.951809856430456e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 4037024,
      "step": 2505
    },
    {
      "epoch": 0.11362864709477354,
      "grad_norm": 0.6040642261505127,
      "learning_rate": 4.951617835420303e-05,
      "loss": 0.2075,
      "num_input_tokens_seen": 4044992,
      "step": 2510
    },
    {
      "epoch": 0.11385499898141652,
      "grad_norm": 0.4458419680595398,
      "learning_rate": 4.951425436339869e-05,
      "loss": 0.2114,
      "num_input_tokens_seen": 4052992,
      "step": 2515
    },
    {
      "epoch": 0.11408135086805948,
      "grad_norm": 0.5544381141662598,
      "learning_rate": 4.9512326592188274e-05,
      "loss": 0.2029,
      "num_input_tokens_seen": 4061824,
      "step": 2520
    },
    {
      "epoch": 0.11430770275470246,
      "grad_norm": 0.8140352368354797,
      "learning_rate": 4.9510395040869054e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 4070496,
      "step": 2525
    },
    {
      "epoch": 0.11453405464134543,
      "grad_norm": 0.919244647026062,
      "learning_rate": 4.9508459709738905e-05,
      "loss": 0.2319,
      "num_input_tokens_seen": 4077632,
      "step": 2530
    },
    {
      "epoch": 0.11476040652798841,
      "grad_norm": 0.9936850070953369,
      "learning_rate": 4.950652059909627e-05,
      "loss": 0.2287,
      "num_input_tokens_seen": 4085696,
      "step": 2535
    },
    {
      "epoch": 0.11498675841463138,
      "grad_norm": 0.323280394077301,
      "learning_rate": 4.95045777092402e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 4093184,
      "step": 2540
    },
    {
      "epoch": 0.11521311030127436,
      "grad_norm": 0.5398260951042175,
      "learning_rate": 4.950263104047031e-05,
      "loss": 0.2108,
      "num_input_tokens_seen": 4101632,
      "step": 2545
    },
    {
      "epoch": 0.11543946218791734,
      "grad_norm": 0.30031806230545044,
      "learning_rate": 4.9500680593086775e-05,
      "loss": 0.1781,
      "num_input_tokens_seen": 4109344,
      "step": 2550
    },
    {
      "epoch": 0.11566581407456031,
      "grad_norm": 0.7931666970252991,
      "learning_rate": 4.94987263673904e-05,
      "loss": 0.2617,
      "num_input_tokens_seen": 4116864,
      "step": 2555
    },
    {
      "epoch": 0.11589216596120329,
      "grad_norm": 0.9025783538818359,
      "learning_rate": 4.949676836368256e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 4124672,
      "step": 2560
    },
    {
      "epoch": 0.11611851784784626,
      "grad_norm": 0.525456964969635,
      "learning_rate": 4.949480658226518e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 4133536,
      "step": 2565
    },
    {
      "epoch": 0.11634486973448924,
      "grad_norm": 0.529723048210144,
      "learning_rate": 4.949284102344082e-05,
      "loss": 0.218,
      "num_input_tokens_seen": 4141408,
      "step": 2570
    },
    {
      "epoch": 0.11657122162113222,
      "grad_norm": 0.6960001587867737,
      "learning_rate": 4.9490871687512565e-05,
      "loss": 0.2164,
      "num_input_tokens_seen": 4149792,
      "step": 2575
    },
    {
      "epoch": 0.11679757350777518,
      "grad_norm": 0.691790759563446,
      "learning_rate": 4.948889857478413e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 4157696,
      "step": 2580
    },
    {
      "epoch": 0.11702392539441817,
      "grad_norm": 0.8422835469245911,
      "learning_rate": 4.948692168555978e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 4165984,
      "step": 2585
    },
    {
      "epoch": 0.11725027728106113,
      "grad_norm": 0.7918756008148193,
      "learning_rate": 4.94849410201444e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 4173120,
      "step": 2590
    },
    {
      "epoch": 0.11747662916770411,
      "grad_norm": 0.7208459973335266,
      "learning_rate": 4.948295657884341e-05,
      "loss": 0.2177,
      "num_input_tokens_seen": 4181376,
      "step": 2595
    },
    {
      "epoch": 0.1177029810543471,
      "grad_norm": 0.43608322739601135,
      "learning_rate": 4.9480968361962835e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 4189440,
      "step": 2600
    },
    {
      "epoch": 0.1177029810543471,
      "eval_loss": 0.200910285115242,
      "eval_runtime": 404.1915,
      "eval_samples_per_second": 97.159,
      "eval_steps_per_second": 24.29,
      "num_input_tokens_seen": 4189440,
      "step": 2600
    },
    {
      "epoch": 0.11792933294099006,
      "grad_norm": 0.8843706250190735,
      "learning_rate": 4.9478976369809305e-05,
      "loss": 0.2428,
      "num_input_tokens_seen": 4197728,
      "step": 2605
    },
    {
      "epoch": 0.11815568482763304,
      "grad_norm": 1.087490200996399,
      "learning_rate": 4.947698060268999e-05,
      "loss": 0.1999,
      "num_input_tokens_seen": 4205504,
      "step": 2610
    },
    {
      "epoch": 0.11838203671427601,
      "grad_norm": 0.7400227189064026,
      "learning_rate": 4.9474981060912665e-05,
      "loss": 0.2369,
      "num_input_tokens_seen": 4213760,
      "step": 2615
    },
    {
      "epoch": 0.11860838860091899,
      "grad_norm": 0.5054681301116943,
      "learning_rate": 4.94729777447857e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 4221600,
      "step": 2620
    },
    {
      "epoch": 0.11883474048756196,
      "grad_norm": 0.6680654883384705,
      "learning_rate": 4.947097065461801e-05,
      "loss": 0.218,
      "num_input_tokens_seen": 4230272,
      "step": 2625
    },
    {
      "epoch": 0.11906109237420494,
      "grad_norm": 0.34956100583076477,
      "learning_rate": 4.9468959790719125e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 4238336,
      "step": 2630
    },
    {
      "epoch": 0.11928744426084792,
      "grad_norm": 0.5864454507827759,
      "learning_rate": 4.9466945153399146e-05,
      "loss": 0.2174,
      "num_input_tokens_seen": 4246080,
      "step": 2635
    },
    {
      "epoch": 0.11951379614749089,
      "grad_norm": 0.6084436178207397,
      "learning_rate": 4.9464926742968755e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 4253952,
      "step": 2640
    },
    {
      "epoch": 0.11974014803413387,
      "grad_norm": 0.7932662963867188,
      "learning_rate": 4.946290455973921e-05,
      "loss": 0.2303,
      "num_input_tokens_seen": 4261824,
      "step": 2645
    },
    {
      "epoch": 0.11996649992077683,
      "grad_norm": 0.6393258571624756,
      "learning_rate": 4.9460878604022365e-05,
      "loss": 0.2003,
      "num_input_tokens_seen": 4270304,
      "step": 2650
    },
    {
      "epoch": 0.12019285180741981,
      "grad_norm": 0.8737585544586182,
      "learning_rate": 4.945884887613065e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 4278432,
      "step": 2655
    },
    {
      "epoch": 0.1204192036940628,
      "grad_norm": 0.5232771039009094,
      "learning_rate": 4.9456815376377055e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 4286464,
      "step": 2660
    },
    {
      "epoch": 0.12064555558070576,
      "grad_norm": 0.3598176836967468,
      "learning_rate": 4.9454778105075195e-05,
      "loss": 0.2106,
      "num_input_tokens_seen": 4294592,
      "step": 2665
    },
    {
      "epoch": 0.12087190746734874,
      "grad_norm": 0.7819569706916809,
      "learning_rate": 4.945273706253924e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 4303456,
      "step": 2670
    },
    {
      "epoch": 0.12109825935399171,
      "grad_norm": 1.0088082551956177,
      "learning_rate": 4.9450692249083925e-05,
      "loss": 0.1985,
      "num_input_tokens_seen": 4312448,
      "step": 2675
    },
    {
      "epoch": 0.12132461124063469,
      "grad_norm": 0.9305062890052795,
      "learning_rate": 4.9448643665024605e-05,
      "loss": 0.2097,
      "num_input_tokens_seen": 4320352,
      "step": 2680
    },
    {
      "epoch": 0.12155096312727767,
      "grad_norm": 0.7798030376434326,
      "learning_rate": 4.944659131067719e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 4328160,
      "step": 2685
    },
    {
      "epoch": 0.12177731501392064,
      "grad_norm": 0.5462347269058228,
      "learning_rate": 4.944453518635818e-05,
      "loss": 0.2146,
      "num_input_tokens_seen": 4335744,
      "step": 2690
    },
    {
      "epoch": 0.12200366690056362,
      "grad_norm": 0.3514373302459717,
      "learning_rate": 4.944247529238465e-05,
      "loss": 0.2054,
      "num_input_tokens_seen": 4343488,
      "step": 2695
    },
    {
      "epoch": 0.12223001878720659,
      "grad_norm": 0.38577505946159363,
      "learning_rate": 4.944041162907427e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 4351488,
      "step": 2700
    },
    {
      "epoch": 0.12245637067384957,
      "grad_norm": 0.8198986053466797,
      "learning_rate": 4.943834419674529e-05,
      "loss": 0.2093,
      "num_input_tokens_seen": 4359552,
      "step": 2705
    },
    {
      "epoch": 0.12268272256049254,
      "grad_norm": 0.3978525400161743,
      "learning_rate": 4.9436272995716506e-05,
      "loss": 0.2352,
      "num_input_tokens_seen": 4367904,
      "step": 2710
    },
    {
      "epoch": 0.12290907444713552,
      "grad_norm": 0.721975564956665,
      "learning_rate": 4.943419802630735e-05,
      "loss": 0.2096,
      "num_input_tokens_seen": 4375712,
      "step": 2715
    },
    {
      "epoch": 0.1231354263337785,
      "grad_norm": 0.4735710620880127,
      "learning_rate": 4.94321192888378e-05,
      "loss": 0.1978,
      "num_input_tokens_seen": 4384160,
      "step": 2720
    },
    {
      "epoch": 0.12336177822042146,
      "grad_norm": 0.33628326654434204,
      "learning_rate": 4.943003678362842e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 4392928,
      "step": 2725
    },
    {
      "epoch": 0.12358813010706445,
      "grad_norm": 0.683290958404541,
      "learning_rate": 4.942795051100036e-05,
      "loss": 0.2188,
      "num_input_tokens_seen": 4400608,
      "step": 2730
    },
    {
      "epoch": 0.12381448199370741,
      "grad_norm": 0.5469914078712463,
      "learning_rate": 4.942586047127536e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 4408512,
      "step": 2735
    },
    {
      "epoch": 0.12404083388035039,
      "grad_norm": 0.6154870390892029,
      "learning_rate": 4.942376666477571e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 4416800,
      "step": 2740
    },
    {
      "epoch": 0.12426718576699337,
      "grad_norm": 0.5154933333396912,
      "learning_rate": 4.9421669091824304e-05,
      "loss": 0.2137,
      "num_input_tokens_seen": 4424576,
      "step": 2745
    },
    {
      "epoch": 0.12449353765363634,
      "grad_norm": 0.4840431809425354,
      "learning_rate": 4.9419567752744634e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 4432320,
      "step": 2750
    },
    {
      "epoch": 0.12471988954027932,
      "grad_norm": 1.4063962697982788,
      "learning_rate": 4.941746264786074e-05,
      "loss": 0.1926,
      "num_input_tokens_seen": 4440704,
      "step": 2755
    },
    {
      "epoch": 0.12494624142692229,
      "grad_norm": 0.5075735449790955,
      "learning_rate": 4.9415353777497254e-05,
      "loss": 0.2354,
      "num_input_tokens_seen": 4449152,
      "step": 2760
    },
    {
      "epoch": 0.12517259331356526,
      "grad_norm": 0.53789883852005,
      "learning_rate": 4.9413241141979394e-05,
      "loss": 0.1928,
      "num_input_tokens_seen": 4457696,
      "step": 2765
    },
    {
      "epoch": 0.12539894520020825,
      "grad_norm": 0.9902108907699585,
      "learning_rate": 4.9411124741632956e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 4465920,
      "step": 2770
    },
    {
      "epoch": 0.12562529708685122,
      "grad_norm": 0.7113134264945984,
      "learning_rate": 4.940900457678431e-05,
      "loss": 0.2015,
      "num_input_tokens_seen": 4473664,
      "step": 2775
    },
    {
      "epoch": 0.12585164897349418,
      "grad_norm": 0.850161612033844,
      "learning_rate": 4.9406880647760425e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 4481088,
      "step": 2780
    },
    {
      "epoch": 0.12607800086013718,
      "grad_norm": 0.5362322330474854,
      "learning_rate": 4.9404752954888824e-05,
      "loss": 0.2174,
      "num_input_tokens_seen": 4488992,
      "step": 2785
    },
    {
      "epoch": 0.12630435274678015,
      "grad_norm": 0.8074328303337097,
      "learning_rate": 4.940262149849762e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 4496928,
      "step": 2790
    },
    {
      "epoch": 0.1265307046334231,
      "grad_norm": 0.7302848100662231,
      "learning_rate": 4.9400486278915526e-05,
      "loss": 0.2043,
      "num_input_tokens_seen": 4504608,
      "step": 2795
    },
    {
      "epoch": 0.1267570565200661,
      "grad_norm": 0.6166640520095825,
      "learning_rate": 4.939834729647181e-05,
      "loss": 0.1858,
      "num_input_tokens_seen": 4512768,
      "step": 2800
    },
    {
      "epoch": 0.1267570565200661,
      "eval_loss": 0.19701775908470154,
      "eval_runtime": 404.1109,
      "eval_samples_per_second": 97.179,
      "eval_steps_per_second": 24.295,
      "num_input_tokens_seen": 4512768,
      "step": 2800
    },
    {
      "epoch": 0.12698340840670908,
      "grad_norm": 0.4750502407550812,
      "learning_rate": 4.9396204551496326e-05,
      "loss": 0.1965,
      "num_input_tokens_seen": 4520704,
      "step": 2805
    },
    {
      "epoch": 0.12720976029335204,
      "grad_norm": 0.394535094499588,
      "learning_rate": 4.939405804431952e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 4528736,
      "step": 2810
    },
    {
      "epoch": 0.127436112179995,
      "grad_norm": 0.5792413949966431,
      "learning_rate": 4.9391907775272414e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 4536608,
      "step": 2815
    },
    {
      "epoch": 0.127662464066638,
      "grad_norm": 0.4738765358924866,
      "learning_rate": 4.9389753744686604e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 4544128,
      "step": 2820
    },
    {
      "epoch": 0.12788881595328097,
      "grad_norm": 0.4544652998447418,
      "learning_rate": 4.938759595289426e-05,
      "loss": 0.2217,
      "num_input_tokens_seen": 4552512,
      "step": 2825
    },
    {
      "epoch": 0.12811516783992394,
      "grad_norm": 0.7706785798072815,
      "learning_rate": 4.938543440022815e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 4560800,
      "step": 2830
    },
    {
      "epoch": 0.12834151972656693,
      "grad_norm": 0.6873086094856262,
      "learning_rate": 4.938326908702161e-05,
      "loss": 0.2141,
      "num_input_tokens_seen": 4568288,
      "step": 2835
    },
    {
      "epoch": 0.1285678716132099,
      "grad_norm": 0.4626162350177765,
      "learning_rate": 4.9381100013608554e-05,
      "loss": 0.2239,
      "num_input_tokens_seen": 4575808,
      "step": 2840
    },
    {
      "epoch": 0.12879422349985287,
      "grad_norm": 0.9327855706214905,
      "learning_rate": 4.9378927180323485e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 4584192,
      "step": 2845
    },
    {
      "epoch": 0.12902057538649583,
      "grad_norm": 0.8944616913795471,
      "learning_rate": 4.937675058750148e-05,
      "loss": 0.2055,
      "num_input_tokens_seen": 4592160,
      "step": 2850
    },
    {
      "epoch": 0.12924692727313883,
      "grad_norm": 0.4216426610946655,
      "learning_rate": 4.937457023547819e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 4600928,
      "step": 2855
    },
    {
      "epoch": 0.1294732791597818,
      "grad_norm": 0.9550073146820068,
      "learning_rate": 4.9372386124589876e-05,
      "loss": 0.1992,
      "num_input_tokens_seen": 4608832,
      "step": 2860
    },
    {
      "epoch": 0.12969963104642476,
      "grad_norm": 0.4180701673030853,
      "learning_rate": 4.937019825517333e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 4616416,
      "step": 2865
    },
    {
      "epoch": 0.12992598293306776,
      "grad_norm": 0.6156051754951477,
      "learning_rate": 4.9368006627565954e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 4624544,
      "step": 2870
    },
    {
      "epoch": 0.13015233481971072,
      "grad_norm": 0.8046523928642273,
      "learning_rate": 4.936581124210573e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 4632160,
      "step": 2875
    },
    {
      "epoch": 0.1303786867063537,
      "grad_norm": 0.4476199448108673,
      "learning_rate": 4.9363612099131216e-05,
      "loss": 0.1981,
      "num_input_tokens_seen": 4640288,
      "step": 2880
    },
    {
      "epoch": 0.1306050385929967,
      "grad_norm": 0.5793073177337646,
      "learning_rate": 4.936140919898155e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 4648640,
      "step": 2885
    },
    {
      "epoch": 0.13083139047963965,
      "grad_norm": 0.28178659081459045,
      "learning_rate": 4.9359202541996426e-05,
      "loss": 0.1982,
      "num_input_tokens_seen": 4656704,
      "step": 2890
    },
    {
      "epoch": 0.13105774236628262,
      "grad_norm": 0.5878974795341492,
      "learning_rate": 4.935699212851616e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 4664288,
      "step": 2895
    },
    {
      "epoch": 0.1312840942529256,
      "grad_norm": 0.5543524026870728,
      "learning_rate": 4.935477795888162e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 4672064,
      "step": 2900
    },
    {
      "epoch": 0.13151044613956858,
      "grad_norm": 0.9407302737236023,
      "learning_rate": 4.935256003343426e-05,
      "loss": 0.2147,
      "num_input_tokens_seen": 4680192,
      "step": 2905
    },
    {
      "epoch": 0.13173679802621155,
      "grad_norm": 0.4934304356575012,
      "learning_rate": 4.93503383525161e-05,
      "loss": 0.2058,
      "num_input_tokens_seen": 4687904,
      "step": 2910
    },
    {
      "epoch": 0.13196314991285452,
      "grad_norm": 0.6764005422592163,
      "learning_rate": 4.934811291646977e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 4696480,
      "step": 2915
    },
    {
      "epoch": 0.1321895017994975,
      "grad_norm": 0.5891165137290955,
      "learning_rate": 4.934588372563845e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 4704480,
      "step": 2920
    },
    {
      "epoch": 0.13241585368614048,
      "grad_norm": 0.7145051956176758,
      "learning_rate": 4.93436507803659e-05,
      "loss": 0.1916,
      "num_input_tokens_seen": 4713120,
      "step": 2925
    },
    {
      "epoch": 0.13264220557278344,
      "grad_norm": 0.44670411944389343,
      "learning_rate": 4.934141408099649e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 4720576,
      "step": 2930
    },
    {
      "epoch": 0.1328685574594264,
      "grad_norm": 0.662880003452301,
      "learning_rate": 4.9339173627875135e-05,
      "loss": 0.2072,
      "num_input_tokens_seen": 4728544,
      "step": 2935
    },
    {
      "epoch": 0.1330949093460694,
      "grad_norm": 1.1516695022583008,
      "learning_rate": 4.9336929421347335e-05,
      "loss": 0.2018,
      "num_input_tokens_seen": 4736768,
      "step": 2940
    },
    {
      "epoch": 0.13332126123271237,
      "grad_norm": 0.9231851100921631,
      "learning_rate": 4.933468146175918e-05,
      "loss": 0.2293,
      "num_input_tokens_seen": 4744864,
      "step": 2945
    },
    {
      "epoch": 0.13354761311935534,
      "grad_norm": 1.1762070655822754,
      "learning_rate": 4.933242974945734e-05,
      "loss": 0.2,
      "num_input_tokens_seen": 4752608,
      "step": 2950
    },
    {
      "epoch": 0.13377396500599834,
      "grad_norm": 0.5314508080482483,
      "learning_rate": 4.933017428478906e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 4760832,
      "step": 2955
    },
    {
      "epoch": 0.1340003168926413,
      "grad_norm": 0.859917938709259,
      "learning_rate": 4.932791506810214e-05,
      "loss": 0.2384,
      "num_input_tokens_seen": 4769376,
      "step": 2960
    },
    {
      "epoch": 0.13422666877928427,
      "grad_norm": 0.49082711338996887,
      "learning_rate": 4.932565209974499e-05,
      "loss": 0.1921,
      "num_input_tokens_seen": 4778304,
      "step": 2965
    },
    {
      "epoch": 0.13445302066592726,
      "grad_norm": 0.633368968963623,
      "learning_rate": 4.93233853800666e-05,
      "loss": 0.197,
      "num_input_tokens_seen": 4786144,
      "step": 2970
    },
    {
      "epoch": 0.13467937255257023,
      "grad_norm": 0.9145513772964478,
      "learning_rate": 4.932111490941651e-05,
      "loss": 0.2145,
      "num_input_tokens_seen": 4793600,
      "step": 2975
    },
    {
      "epoch": 0.1349057244392132,
      "grad_norm": 0.6927295923233032,
      "learning_rate": 4.9318840688144876e-05,
      "loss": 0.1788,
      "num_input_tokens_seen": 4801984,
      "step": 2980
    },
    {
      "epoch": 0.13513207632585617,
      "grad_norm": 0.6536099314689636,
      "learning_rate": 4.9316562716602387e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 4810112,
      "step": 2985
    },
    {
      "epoch": 0.13535842821249916,
      "grad_norm": 0.6198363304138184,
      "learning_rate": 4.9314280995140346e-05,
      "loss": 0.1857,
      "num_input_tokens_seen": 4817952,
      "step": 2990
    },
    {
      "epoch": 0.13558478009914213,
      "grad_norm": 0.4969932436943054,
      "learning_rate": 4.931199552411063e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 4825504,
      "step": 2995
    },
    {
      "epoch": 0.1358111319857851,
      "grad_norm": 0.9639872908592224,
      "learning_rate": 4.930970630386568e-05,
      "loss": 0.1975,
      "num_input_tokens_seen": 4833792,
      "step": 3000
    },
    {
      "epoch": 0.1358111319857851,
      "eval_loss": 0.1939389407634735,
      "eval_runtime": 404.9971,
      "eval_samples_per_second": 96.966,
      "eval_steps_per_second": 24.242,
      "num_input_tokens_seen": 4833792,
      "step": 3000
    },
    {
      "epoch": 0.1360374838724281,
      "grad_norm": 0.6876811981201172,
      "learning_rate": 4.9307413334758524e-05,
      "loss": 0.2016,
      "num_input_tokens_seen": 4842144,
      "step": 3005
    },
    {
      "epoch": 0.13626383575907106,
      "grad_norm": 0.34531137347221375,
      "learning_rate": 4.930511661714276e-05,
      "loss": 0.1997,
      "num_input_tokens_seen": 4850368,
      "step": 3010
    },
    {
      "epoch": 0.13649018764571402,
      "grad_norm": 0.7427001595497131,
      "learning_rate": 4.9302816151372576e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 4858656,
      "step": 3015
    },
    {
      "epoch": 0.136716539532357,
      "grad_norm": 1.4352924823760986,
      "learning_rate": 4.930051193780274e-05,
      "loss": 0.2199,
      "num_input_tokens_seen": 4866912,
      "step": 3020
    },
    {
      "epoch": 0.13694289141899998,
      "grad_norm": 0.8234250545501709,
      "learning_rate": 4.929820397678858e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 4874464,
      "step": 3025
    },
    {
      "epoch": 0.13716924330564295,
      "grad_norm": 0.8062819242477417,
      "learning_rate": 4.9295892268686015e-05,
      "loss": 0.2147,
      "num_input_tokens_seen": 4883168,
      "step": 3030
    },
    {
      "epoch": 0.13739559519228592,
      "grad_norm": 0.8986985683441162,
      "learning_rate": 4.9293576813851536e-05,
      "loss": 0.2336,
      "num_input_tokens_seen": 4892064,
      "step": 3035
    },
    {
      "epoch": 0.1376219470789289,
      "grad_norm": 0.5436599850654602,
      "learning_rate": 4.929125761264223e-05,
      "loss": 0.2129,
      "num_input_tokens_seen": 4900736,
      "step": 3040
    },
    {
      "epoch": 0.13784829896557188,
      "grad_norm": 0.3242480158805847,
      "learning_rate": 4.928893466541573e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 4908672,
      "step": 3045
    },
    {
      "epoch": 0.13807465085221485,
      "grad_norm": 0.6435105204582214,
      "learning_rate": 4.928660797253027e-05,
      "loss": 0.2179,
      "num_input_tokens_seen": 4917376,
      "step": 3050
    },
    {
      "epoch": 0.13830100273885784,
      "grad_norm": 5.2546916007995605,
      "learning_rate": 4.928427753434467e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 4925216,
      "step": 3055
    },
    {
      "epoch": 0.1385273546255008,
      "grad_norm": 0.7332722544670105,
      "learning_rate": 4.9281943351218286e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 4933504,
      "step": 3060
    },
    {
      "epoch": 0.13875370651214378,
      "grad_norm": 0.3625313639640808,
      "learning_rate": 4.9279605423511095e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 4941792,
      "step": 3065
    },
    {
      "epoch": 0.13898005839878674,
      "grad_norm": 0.7468270659446716,
      "learning_rate": 4.927726375158363e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 4949568,
      "step": 3070
    },
    {
      "epoch": 0.13920641028542974,
      "grad_norm": 0.39308199286460876,
      "learning_rate": 4.9274918335797004e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 4957600,
      "step": 3075
    },
    {
      "epoch": 0.1394327621720727,
      "grad_norm": 0.48098790645599365,
      "learning_rate": 4.927256917651292e-05,
      "loss": 0.2008,
      "num_input_tokens_seen": 4966400,
      "step": 3080
    },
    {
      "epoch": 0.13965911405871567,
      "grad_norm": 0.36692720651626587,
      "learning_rate": 4.927021627409364e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 4974048,
      "step": 3085
    },
    {
      "epoch": 0.13988546594535867,
      "grad_norm": 0.5265671610832214,
      "learning_rate": 4.9267859628902005e-05,
      "loss": 0.204,
      "num_input_tokens_seen": 4982464,
      "step": 3090
    },
    {
      "epoch": 0.14011181783200163,
      "grad_norm": 0.5953001976013184,
      "learning_rate": 4.9265499241301454e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 4990272,
      "step": 3095
    },
    {
      "epoch": 0.1403381697186446,
      "grad_norm": 0.3143351972103119,
      "learning_rate": 4.926313511165598e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 4998208,
      "step": 3100
    },
    {
      "epoch": 0.14056452160528757,
      "grad_norm": 0.4539456367492676,
      "learning_rate": 4.926076724033016e-05,
      "loss": 0.2065,
      "num_input_tokens_seen": 5006240,
      "step": 3105
    },
    {
      "epoch": 0.14079087349193056,
      "grad_norm": 0.5834905505180359,
      "learning_rate": 4.9258395627689146e-05,
      "loss": 0.2022,
      "num_input_tokens_seen": 5013856,
      "step": 3110
    },
    {
      "epoch": 0.14101722537857353,
      "grad_norm": 0.5317695736885071,
      "learning_rate": 4.925602027409868e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 5021216,
      "step": 3115
    },
    {
      "epoch": 0.1412435772652165,
      "grad_norm": 0.7834122776985168,
      "learning_rate": 4.925364117992507e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 5029088,
      "step": 3120
    },
    {
      "epoch": 0.1414699291518595,
      "grad_norm": 0.617635190486908,
      "learning_rate": 4.92512583455352e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 5037600,
      "step": 3125
    },
    {
      "epoch": 0.14169628103850246,
      "grad_norm": 1.0549153089523315,
      "learning_rate": 4.9248871771296536e-05,
      "loss": 0.2261,
      "num_input_tokens_seen": 5045600,
      "step": 3130
    },
    {
      "epoch": 0.14192263292514543,
      "grad_norm": 0.5020622611045837,
      "learning_rate": 4.924648145757711e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 5053408,
      "step": 3135
    },
    {
      "epoch": 0.1421489848117884,
      "grad_norm": 0.5925608277320862,
      "learning_rate": 4.924408740474554e-05,
      "loss": 0.2148,
      "num_input_tokens_seen": 5061440,
      "step": 3140
    },
    {
      "epoch": 0.1423753366984314,
      "grad_norm": 0.7044655680656433,
      "learning_rate": 4.924168961317103e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 5068992,
      "step": 3145
    },
    {
      "epoch": 0.14260168858507435,
      "grad_norm": 0.7974086999893188,
      "learning_rate": 4.9239288083223334e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 5076896,
      "step": 3150
    },
    {
      "epoch": 0.14282804047171732,
      "grad_norm": 0.6508958339691162,
      "learning_rate": 4.9236882815272803e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 5085504,
      "step": 3155
    },
    {
      "epoch": 0.14305439235836032,
      "grad_norm": 0.3287821412086487,
      "learning_rate": 4.9234473809690365e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 5093120,
      "step": 3160
    },
    {
      "epoch": 0.14328074424500328,
      "grad_norm": 0.43852439522743225,
      "learning_rate": 4.923206106684752e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 5101152,
      "step": 3165
    },
    {
      "epoch": 0.14350709613164625,
      "grad_norm": 0.5349432826042175,
      "learning_rate": 4.922964458711634e-05,
      "loss": 0.2165,
      "num_input_tokens_seen": 5109216,
      "step": 3170
    },
    {
      "epoch": 0.14373344801828924,
      "grad_norm": 0.6224165558815002,
      "learning_rate": 4.9227224370869474e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 5117760,
      "step": 3175
    },
    {
      "epoch": 0.1439597999049322,
      "grad_norm": 0.7693657279014587,
      "learning_rate": 4.9224800418480155e-05,
      "loss": 0.1962,
      "num_input_tokens_seen": 5125568,
      "step": 3180
    },
    {
      "epoch": 0.14418615179157518,
      "grad_norm": 0.5448944568634033,
      "learning_rate": 4.9222372730322176e-05,
      "loss": 0.1845,
      "num_input_tokens_seen": 5133472,
      "step": 3185
    },
    {
      "epoch": 0.14441250367821815,
      "grad_norm": 0.4518173635005951,
      "learning_rate": 4.921994130676993e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 5141184,
      "step": 3190
    },
    {
      "epoch": 0.14463885556486114,
      "grad_norm": 0.7522606253623962,
      "learning_rate": 4.9217506148198366e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 5149504,
      "step": 3195
    },
    {
      "epoch": 0.1448652074515041,
      "grad_norm": 0.7602168321609497,
      "learning_rate": 4.921506725498302e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 5157664,
      "step": 3200
    },
    {
      "epoch": 0.1448652074515041,
      "eval_loss": 0.19143176078796387,
      "eval_runtime": 403.9734,
      "eval_samples_per_second": 97.212,
      "eval_steps_per_second": 24.304,
      "num_input_tokens_seen": 5157664,
      "step": 3200
    },
    {
      "epoch": 0.14509155933814707,
      "grad_norm": 0.9246489405632019,
      "learning_rate": 4.9212624627499994e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 5165408,
      "step": 3205
    },
    {
      "epoch": 0.14531791122479007,
      "grad_norm": 0.5462193489074707,
      "learning_rate": 4.921017826612597e-05,
      "loss": 0.2235,
      "num_input_tokens_seen": 5173632,
      "step": 3210
    },
    {
      "epoch": 0.14554426311143304,
      "grad_norm": 0.71478670835495,
      "learning_rate": 4.9207728171238223e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 5181760,
      "step": 3215
    },
    {
      "epoch": 0.145770614998076,
      "grad_norm": 0.6106526851654053,
      "learning_rate": 4.920527434321458e-05,
      "loss": 0.238,
      "num_input_tokens_seen": 5189888,
      "step": 3220
    },
    {
      "epoch": 0.14599696688471897,
      "grad_norm": 1.425859808921814,
      "learning_rate": 4.920281678243345e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 5197568,
      "step": 3225
    },
    {
      "epoch": 0.14622331877136197,
      "grad_norm": 0.8500978946685791,
      "learning_rate": 4.920035548927381e-05,
      "loss": 0.2281,
      "num_input_tokens_seen": 5204992,
      "step": 3230
    },
    {
      "epoch": 0.14644967065800493,
      "grad_norm": 0.8247193098068237,
      "learning_rate": 4.919789046411525e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 5213184,
      "step": 3235
    },
    {
      "epoch": 0.1466760225446479,
      "grad_norm": 0.42784813046455383,
      "learning_rate": 4.919542170733787e-05,
      "loss": 0.202,
      "num_input_tokens_seen": 5220864,
      "step": 3240
    },
    {
      "epoch": 0.1469023744312909,
      "grad_norm": 0.4381972849369049,
      "learning_rate": 4.919294921932242e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 5229920,
      "step": 3245
    },
    {
      "epoch": 0.14712872631793386,
      "grad_norm": 0.8194603323936462,
      "learning_rate": 4.919047300045016e-05,
      "loss": 0.2107,
      "num_input_tokens_seen": 5237472,
      "step": 3250
    },
    {
      "epoch": 0.14735507820457683,
      "grad_norm": 0.45081618428230286,
      "learning_rate": 4.918799305110299e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 5245792,
      "step": 3255
    },
    {
      "epoch": 0.14758143009121982,
      "grad_norm": 0.4970557689666748,
      "learning_rate": 4.918550937166331e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 5253888,
      "step": 3260
    },
    {
      "epoch": 0.1478077819778628,
      "grad_norm": 0.4951764643192291,
      "learning_rate": 4.918302196251415e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 5261344,
      "step": 3265
    },
    {
      "epoch": 0.14803413386450576,
      "grad_norm": 0.7598392367362976,
      "learning_rate": 4.91805308240391e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 5270048,
      "step": 3270
    },
    {
      "epoch": 0.14826048575114872,
      "grad_norm": 0.5684120655059814,
      "learning_rate": 4.9178035956622326e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 5278272,
      "step": 3275
    },
    {
      "epoch": 0.14848683763779172,
      "grad_norm": 0.4952169358730316,
      "learning_rate": 4.917553736064857e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 5286368,
      "step": 3280
    },
    {
      "epoch": 0.14871318952443469,
      "grad_norm": 0.507277250289917,
      "learning_rate": 4.917303503650314e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 5294624,
      "step": 3285
    },
    {
      "epoch": 0.14893954141107765,
      "grad_norm": 0.5847182869911194,
      "learning_rate": 4.917052898457194e-05,
      "loss": 0.199,
      "num_input_tokens_seen": 5303008,
      "step": 3290
    },
    {
      "epoch": 0.14916589329772065,
      "grad_norm": 0.9368883371353149,
      "learning_rate": 4.916801920524141e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 5311360,
      "step": 3295
    },
    {
      "epoch": 0.14939224518436361,
      "grad_norm": 0.841032087802887,
      "learning_rate": 4.916550569889862e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 5318976,
      "step": 3300
    },
    {
      "epoch": 0.14961859707100658,
      "grad_norm": 0.8227860927581787,
      "learning_rate": 4.916298846593116e-05,
      "loss": 0.215,
      "num_input_tokens_seen": 5327392,
      "step": 3305
    },
    {
      "epoch": 0.14984494895764955,
      "grad_norm": 0.40393927693367004,
      "learning_rate": 4.916046750672722e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 5335040,
      "step": 3310
    },
    {
      "epoch": 0.15007130084429254,
      "grad_norm": 0.6729105114936829,
      "learning_rate": 4.915794282167559e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 5343136,
      "step": 3315
    },
    {
      "epoch": 0.1502976527309355,
      "grad_norm": 0.6322305798530579,
      "learning_rate": 4.915541441116558e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 5351424,
      "step": 3320
    },
    {
      "epoch": 0.15052400461757848,
      "grad_norm": 0.5620431900024414,
      "learning_rate": 4.915288227558711e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 5359776,
      "step": 3325
    },
    {
      "epoch": 0.15075035650422147,
      "grad_norm": 1.1461193561553955,
      "learning_rate": 4.915034641533066e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 5367360,
      "step": 3330
    },
    {
      "epoch": 0.15097670839086444,
      "grad_norm": 0.5240876078605652,
      "learning_rate": 4.914780683078731e-05,
      "loss": 0.1956,
      "num_input_tokens_seen": 5375456,
      "step": 3335
    },
    {
      "epoch": 0.1512030602775074,
      "grad_norm": 0.9230265021324158,
      "learning_rate": 4.9145263522348695e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 5383456,
      "step": 3340
    },
    {
      "epoch": 0.1514294121641504,
      "grad_norm": 0.577843189239502,
      "learning_rate": 4.9142716490407e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 5391264,
      "step": 3345
    },
    {
      "epoch": 0.15165576405079337,
      "grad_norm": 0.4743180572986603,
      "learning_rate": 4.914016573535504e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 5398976,
      "step": 3350
    },
    {
      "epoch": 0.15188211593743633,
      "grad_norm": 0.4403809905052185,
      "learning_rate": 4.9137611257586154e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 5406752,
      "step": 3355
    },
    {
      "epoch": 0.1521084678240793,
      "grad_norm": 0.28023287653923035,
      "learning_rate": 4.9135053057494274e-05,
      "loss": 0.2071,
      "num_input_tokens_seen": 5413952,
      "step": 3360
    },
    {
      "epoch": 0.1523348197107223,
      "grad_norm": 0.9200780391693115,
      "learning_rate": 4.913249113547392e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 5422048,
      "step": 3365
    },
    {
      "epoch": 0.15256117159736526,
      "grad_norm": 0.8312341570854187,
      "learning_rate": 4.912992549192016e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 5429984,
      "step": 3370
    },
    {
      "epoch": 0.15278752348400823,
      "grad_norm": 0.9368993639945984,
      "learning_rate": 4.9127356127228665e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 5438208,
      "step": 3375
    },
    {
      "epoch": 0.15301387537065123,
      "grad_norm": 0.37554630637168884,
      "learning_rate": 4.912478304179564e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 5446240,
      "step": 3380
    },
    {
      "epoch": 0.1532402272572942,
      "grad_norm": 0.7571684122085571,
      "learning_rate": 4.9122206236017896e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 5454688,
      "step": 3385
    },
    {
      "epoch": 0.15346657914393716,
      "grad_norm": 0.6122147440910339,
      "learning_rate": 4.911962571029282e-05,
      "loss": 0.2061,
      "num_input_tokens_seen": 5462336,
      "step": 3390
    },
    {
      "epoch": 0.15369293103058013,
      "grad_norm": 0.49280932545661926,
      "learning_rate": 4.9117041465018353e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 5470592,
      "step": 3395
    },
    {
      "epoch": 0.15391928291722312,
      "grad_norm": 1.0112069845199585,
      "learning_rate": 4.911445350059302e-05,
      "loss": 0.1926,
      "num_input_tokens_seen": 5479168,
      "step": 3400
    },
    {
      "epoch": 0.15391928291722312,
      "eval_loss": 0.18914565443992615,
      "eval_runtime": 404.4256,
      "eval_samples_per_second": 97.103,
      "eval_steps_per_second": 24.276,
      "num_input_tokens_seen": 5479168,
      "step": 3400
    },
    {
      "epoch": 0.1541456348038661,
      "grad_norm": 0.6344249844551086,
      "learning_rate": 4.9111861817415905e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 5486816,
      "step": 3405
    },
    {
      "epoch": 0.15437198669050906,
      "grad_norm": 0.291007399559021,
      "learning_rate": 4.91092664158867e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 5494944,
      "step": 3410
    },
    {
      "epoch": 0.15459833857715205,
      "grad_norm": 0.8467265367507935,
      "learning_rate": 4.910666729640563e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 5502560,
      "step": 3415
    },
    {
      "epoch": 0.15482469046379502,
      "grad_norm": 0.5598991513252258,
      "learning_rate": 4.910406445937353e-05,
      "loss": 0.2265,
      "num_input_tokens_seen": 5510496,
      "step": 3420
    },
    {
      "epoch": 0.15505104235043798,
      "grad_norm": 1.0044292211532593,
      "learning_rate": 4.9101457905191774e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 5519104,
      "step": 3425
    },
    {
      "epoch": 0.15527739423708098,
      "grad_norm": 0.5882290005683899,
      "learning_rate": 4.909884763426233e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 5527328,
      "step": 3430
    },
    {
      "epoch": 0.15550374612372395,
      "grad_norm": 0.955317497253418,
      "learning_rate": 4.9096233646987736e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 5535040,
      "step": 3435
    },
    {
      "epoch": 0.1557300980103669,
      "grad_norm": 0.8146306276321411,
      "learning_rate": 4.9093615943771104e-05,
      "loss": 0.207,
      "num_input_tokens_seen": 5543744,
      "step": 3440
    },
    {
      "epoch": 0.15595644989700988,
      "grad_norm": 0.6254796981811523,
      "learning_rate": 4.909099452501611e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 5551904,
      "step": 3445
    },
    {
      "epoch": 0.15618280178365287,
      "grad_norm": 0.4362823963165283,
      "learning_rate": 4.908836939112702e-05,
      "loss": 0.202,
      "num_input_tokens_seen": 5559840,
      "step": 3450
    },
    {
      "epoch": 0.15640915367029584,
      "grad_norm": 0.6585093140602112,
      "learning_rate": 4.908574054250865e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 5568128,
      "step": 3455
    },
    {
      "epoch": 0.1566355055569388,
      "grad_norm": 0.5433496236801147,
      "learning_rate": 4.9083107979566414e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 5576128,
      "step": 3460
    },
    {
      "epoch": 0.1568618574435818,
      "grad_norm": 0.874411404132843,
      "learning_rate": 4.908047170270628e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 5583808,
      "step": 3465
    },
    {
      "epoch": 0.15708820933022477,
      "grad_norm": 0.4954471290111542,
      "learning_rate": 4.9077831712334784e-05,
      "loss": 0.182,
      "num_input_tokens_seen": 5592224,
      "step": 3470
    },
    {
      "epoch": 0.15731456121686774,
      "grad_norm": 0.4294126629829407,
      "learning_rate": 4.907518800885907e-05,
      "loss": 0.1927,
      "num_input_tokens_seen": 5599776,
      "step": 3475
    },
    {
      "epoch": 0.1575409131035107,
      "grad_norm": 0.3452622890472412,
      "learning_rate": 4.907254059268681e-05,
      "loss": 0.2081,
      "num_input_tokens_seen": 5607552,
      "step": 3480
    },
    {
      "epoch": 0.1577672649901537,
      "grad_norm": 0.32880476117134094,
      "learning_rate": 4.906988946422628e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 5615456,
      "step": 3485
    },
    {
      "epoch": 0.15799361687679667,
      "grad_norm": 0.5076728463172913,
      "learning_rate": 4.9067234623886315e-05,
      "loss": 0.1892,
      "num_input_tokens_seen": 5623648,
      "step": 3490
    },
    {
      "epoch": 0.15821996876343963,
      "grad_norm": 0.5868281126022339,
      "learning_rate": 4.9064576072076316e-05,
      "loss": 0.194,
      "num_input_tokens_seen": 5632256,
      "step": 3495
    },
    {
      "epoch": 0.15844632065008263,
      "grad_norm": 0.4209249019622803,
      "learning_rate": 4.906191380920628e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 5640160,
      "step": 3500
    },
    {
      "epoch": 0.1586726725367256,
      "grad_norm": 0.4682184159755707,
      "learning_rate": 4.905924783568675e-05,
      "loss": 0.1881,
      "num_input_tokens_seen": 5649696,
      "step": 3505
    },
    {
      "epoch": 0.15889902442336856,
      "grad_norm": 0.40222665667533875,
      "learning_rate": 4.905657815192886e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 5657152,
      "step": 3510
    },
    {
      "epoch": 0.15912537631001156,
      "grad_norm": 0.4866090416908264,
      "learning_rate": 4.90539047583443e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 5664864,
      "step": 3515
    },
    {
      "epoch": 0.15935172819665452,
      "grad_norm": 0.4357738196849823,
      "learning_rate": 4.905122765534534e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 5672384,
      "step": 3520
    },
    {
      "epoch": 0.1595780800832975,
      "grad_norm": 0.9197971224784851,
      "learning_rate": 4.9048546843344846e-05,
      "loss": 0.2092,
      "num_input_tokens_seen": 5680384,
      "step": 3525
    },
    {
      "epoch": 0.15980443196994046,
      "grad_norm": 0.5121914148330688,
      "learning_rate": 4.9045862322756206e-05,
      "loss": 0.2154,
      "num_input_tokens_seen": 5688256,
      "step": 3530
    },
    {
      "epoch": 0.16003078385658345,
      "grad_norm": 0.559961199760437,
      "learning_rate": 4.904317409399342e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 5695904,
      "step": 3535
    },
    {
      "epoch": 0.16025713574322642,
      "grad_norm": 0.2906869649887085,
      "learning_rate": 4.904048215747104e-05,
      "loss": 0.2115,
      "num_input_tokens_seen": 5704864,
      "step": 3540
    },
    {
      "epoch": 0.1604834876298694,
      "grad_norm": 1.2967097759246826,
      "learning_rate": 4.90377865136042e-05,
      "loss": 0.2333,
      "num_input_tokens_seen": 5713216,
      "step": 3545
    },
    {
      "epoch": 0.16070983951651238,
      "grad_norm": 0.4041247069835663,
      "learning_rate": 4.90350871628086e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 5720928,
      "step": 3550
    },
    {
      "epoch": 0.16093619140315535,
      "grad_norm": 0.5306416153907776,
      "learning_rate": 4.903238410550052e-05,
      "loss": 0.1988,
      "num_input_tokens_seen": 5729376,
      "step": 3555
    },
    {
      "epoch": 0.16116254328979832,
      "grad_norm": 0.42004072666168213,
      "learning_rate": 4.90296773420968e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 5738016,
      "step": 3560
    },
    {
      "epoch": 0.16138889517644128,
      "grad_norm": 0.38530901074409485,
      "learning_rate": 4.902696687301486e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 5746784,
      "step": 3565
    },
    {
      "epoch": 0.16161524706308428,
      "grad_norm": 0.38320866227149963,
      "learning_rate": 4.902425269867268e-05,
      "loss": 0.19,
      "num_input_tokens_seen": 5754624,
      "step": 3570
    },
    {
      "epoch": 0.16184159894972724,
      "grad_norm": 0.41925325989723206,
      "learning_rate": 4.902153481948883e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 5763008,
      "step": 3575
    },
    {
      "epoch": 0.1620679508363702,
      "grad_norm": 0.604088306427002,
      "learning_rate": 4.901881323588244e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 5771072,
      "step": 3580
    },
    {
      "epoch": 0.1622943027230132,
      "grad_norm": 0.5824614763259888,
      "learning_rate": 4.90160879482732e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 5780128,
      "step": 3585
    },
    {
      "epoch": 0.16252065460965617,
      "grad_norm": 0.7316604852676392,
      "learning_rate": 4.9013358957081405e-05,
      "loss": 0.2087,
      "num_input_tokens_seen": 5787744,
      "step": 3590
    },
    {
      "epoch": 0.16274700649629914,
      "grad_norm": 0.8024495840072632,
      "learning_rate": 4.901062626272789e-05,
      "loss": 0.21,
      "num_input_tokens_seen": 5795712,
      "step": 3595
    },
    {
      "epoch": 0.16297335838294214,
      "grad_norm": 0.49098509550094604,
      "learning_rate": 4.900788986563406e-05,
      "loss": 0.2202,
      "num_input_tokens_seen": 5803680,
      "step": 3600
    },
    {
      "epoch": 0.16297335838294214,
      "eval_loss": 0.18853440880775452,
      "eval_runtime": 404.1477,
      "eval_samples_per_second": 97.17,
      "eval_steps_per_second": 24.293,
      "num_input_tokens_seen": 5803680,
      "step": 3600
    },
    {
      "epoch": 0.1631997102695851,
      "grad_norm": 0.8053663372993469,
      "learning_rate": 4.9005149766221915e-05,
      "loss": 0.1936,
      "num_input_tokens_seen": 5812448,
      "step": 3605
    },
    {
      "epoch": 0.16342606215622807,
      "grad_norm": 0.42286840081214905,
      "learning_rate": 4.9002405964914e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 5819904,
      "step": 3610
    },
    {
      "epoch": 0.16365241404287104,
      "grad_norm": 0.6240796446800232,
      "learning_rate": 4.899965846213346e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 5827360,
      "step": 3615
    },
    {
      "epoch": 0.16387876592951403,
      "grad_norm": 0.6218278408050537,
      "learning_rate": 4.899690725830399e-05,
      "loss": 0.1785,
      "num_input_tokens_seen": 5835712,
      "step": 3620
    },
    {
      "epoch": 0.164105117816157,
      "grad_norm": 0.5430747270584106,
      "learning_rate": 4.899415235384985e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 5843968,
      "step": 3625
    },
    {
      "epoch": 0.16433146970279996,
      "grad_norm": 0.9485837817192078,
      "learning_rate": 4.899139374919589e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 5851936,
      "step": 3630
    },
    {
      "epoch": 0.16455782158944296,
      "grad_norm": 1.0010871887207031,
      "learning_rate": 4.898863144476752e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 5859904,
      "step": 3635
    },
    {
      "epoch": 0.16478417347608593,
      "grad_norm": 0.8807037472724915,
      "learning_rate": 4.898586544099072e-05,
      "loss": 0.2111,
      "num_input_tokens_seen": 5867456,
      "step": 3640
    },
    {
      "epoch": 0.1650105253627289,
      "grad_norm": 0.7284684777259827,
      "learning_rate": 4.898309573829204e-05,
      "loss": 0.2008,
      "num_input_tokens_seen": 5875232,
      "step": 3645
    },
    {
      "epoch": 0.16523687724937186,
      "grad_norm": 0.9543615579605103,
      "learning_rate": 4.898032233709862e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 5883648,
      "step": 3650
    },
    {
      "epoch": 0.16546322913601486,
      "grad_norm": 0.9493927359580994,
      "learning_rate": 4.8977545237838123e-05,
      "loss": 0.1911,
      "num_input_tokens_seen": 5891136,
      "step": 3655
    },
    {
      "epoch": 0.16568958102265782,
      "grad_norm": 1.2087173461914062,
      "learning_rate": 4.8974764440938836e-05,
      "loss": 0.2288,
      "num_input_tokens_seen": 5898880,
      "step": 3660
    },
    {
      "epoch": 0.1659159329093008,
      "grad_norm": 0.403558611869812,
      "learning_rate": 4.897197994682959e-05,
      "loss": 0.1821,
      "num_input_tokens_seen": 5906784,
      "step": 3665
    },
    {
      "epoch": 0.16614228479594378,
      "grad_norm": 0.6259910464286804,
      "learning_rate": 4.8969191755939786e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 5914432,
      "step": 3670
    },
    {
      "epoch": 0.16636863668258675,
      "grad_norm": 0.5279462933540344,
      "learning_rate": 4.8966399868699396e-05,
      "loss": 0.2242,
      "num_input_tokens_seen": 5921952,
      "step": 3675
    },
    {
      "epoch": 0.16659498856922972,
      "grad_norm": 0.37999364733695984,
      "learning_rate": 4.8963604285538965e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 5929920,
      "step": 3680
    },
    {
      "epoch": 0.1668213404558727,
      "grad_norm": 1.3204509019851685,
      "learning_rate": 4.8960805006889604e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 5937664,
      "step": 3685
    },
    {
      "epoch": 0.16704769234251568,
      "grad_norm": 0.9645914435386658,
      "learning_rate": 4.8958002033183004e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 5945792,
      "step": 3690
    },
    {
      "epoch": 0.16727404422915865,
      "grad_norm": 0.4590088725090027,
      "learning_rate": 4.8955195364851414e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 5953888,
      "step": 3695
    },
    {
      "epoch": 0.16750039611580161,
      "grad_norm": 0.4175783395767212,
      "learning_rate": 4.895238500232766e-05,
      "loss": 0.1857,
      "num_input_tokens_seen": 5961952,
      "step": 3700
    },
    {
      "epoch": 0.1677267480024446,
      "grad_norm": 0.5089616179466248,
      "learning_rate": 4.8949570946045143e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 5969952,
      "step": 3705
    },
    {
      "epoch": 0.16795309988908758,
      "grad_norm": 0.9368423223495483,
      "learning_rate": 4.89467531964378e-05,
      "loss": 0.2184,
      "num_input_tokens_seen": 5977952,
      "step": 3710
    },
    {
      "epoch": 0.16817945177573054,
      "grad_norm": 0.596297562122345,
      "learning_rate": 4.894393175394019e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 5986208,
      "step": 3715
    },
    {
      "epoch": 0.16840580366237354,
      "grad_norm": 0.3654022216796875,
      "learning_rate": 4.8941106618987406e-05,
      "loss": 0.1909,
      "num_input_tokens_seen": 5994560,
      "step": 3720
    },
    {
      "epoch": 0.1686321555490165,
      "grad_norm": 0.801050066947937,
      "learning_rate": 4.893827779201512e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 6002656,
      "step": 3725
    },
    {
      "epoch": 0.16885850743565947,
      "grad_norm": 0.45969071984291077,
      "learning_rate": 4.893544527345957e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 6009888,
      "step": 3730
    },
    {
      "epoch": 0.16908485932230244,
      "grad_norm": 0.4480396509170532,
      "learning_rate": 4.8932609063757563e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 6018144,
      "step": 3735
    },
    {
      "epoch": 0.16931121120894543,
      "grad_norm": 0.3092934191226959,
      "learning_rate": 4.8929769163346484e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 6026368,
      "step": 3740
    },
    {
      "epoch": 0.1695375630955884,
      "grad_norm": 0.33272668719291687,
      "learning_rate": 4.892692557266429e-05,
      "loss": 0.1913,
      "num_input_tokens_seen": 6034528,
      "step": 3745
    },
    {
      "epoch": 0.16976391498223137,
      "grad_norm": 0.7087588310241699,
      "learning_rate": 4.8924078292149464e-05,
      "loss": 0.1944,
      "num_input_tokens_seen": 6042912,
      "step": 3750
    },
    {
      "epoch": 0.16999026686887436,
      "grad_norm": 1.4333171844482422,
      "learning_rate": 4.892122732224114e-05,
      "loss": 0.1977,
      "num_input_tokens_seen": 6051232,
      "step": 3755
    },
    {
      "epoch": 0.17021661875551733,
      "grad_norm": 0.5771207213401794,
      "learning_rate": 4.8918372663378944e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 6059456,
      "step": 3760
    },
    {
      "epoch": 0.1704429706421603,
      "grad_norm": 0.564551591873169,
      "learning_rate": 4.89155143160031e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 6067904,
      "step": 3765
    },
    {
      "epoch": 0.1706693225288033,
      "grad_norm": 0.4638155698776245,
      "learning_rate": 4.891265228055441e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 6076320,
      "step": 3770
    },
    {
      "epoch": 0.17089567441544626,
      "grad_norm": 0.9677026271820068,
      "learning_rate": 4.890978655747424e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 6084224,
      "step": 3775
    },
    {
      "epoch": 0.17112202630208923,
      "grad_norm": 0.3484242260456085,
      "learning_rate": 4.89069171472045e-05,
      "loss": 0.2273,
      "num_input_tokens_seen": 6091776,
      "step": 3780
    },
    {
      "epoch": 0.1713483781887322,
      "grad_norm": 0.8121957778930664,
      "learning_rate": 4.890404405018772e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 6100320,
      "step": 3785
    },
    {
      "epoch": 0.1715747300753752,
      "grad_norm": 0.5536677837371826,
      "learning_rate": 4.8901167266866934e-05,
      "loss": 0.2191,
      "num_input_tokens_seen": 6108384,
      "step": 3790
    },
    {
      "epoch": 0.17180108196201815,
      "grad_norm": 0.6253343820571899,
      "learning_rate": 4.88982867976858e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 6116832,
      "step": 3795
    },
    {
      "epoch": 0.17202743384866112,
      "grad_norm": 0.5535609126091003,
      "learning_rate": 4.889540264308852e-05,
      "loss": 0.2021,
      "num_input_tokens_seen": 6125472,
      "step": 3800
    },
    {
      "epoch": 0.17202743384866112,
      "eval_loss": 0.18630996346473694,
      "eval_runtime": 404.2774,
      "eval_samples_per_second": 97.139,
      "eval_steps_per_second": 24.285,
      "num_input_tokens_seen": 6125472,
      "step": 3800
    },
    {
      "epoch": 0.17225378573530412,
      "grad_norm": 0.7655861377716064,
      "learning_rate": 4.889251480351986e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 6132992,
      "step": 3805
    },
    {
      "epoch": 0.17248013762194708,
      "grad_norm": 0.8657500743865967,
      "learning_rate": 4.888962327942517e-05,
      "loss": 0.2062,
      "num_input_tokens_seen": 6140928,
      "step": 3810
    },
    {
      "epoch": 0.17270648950859005,
      "grad_norm": 0.7537941932678223,
      "learning_rate": 4.8886728071250356e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 6149024,
      "step": 3815
    },
    {
      "epoch": 0.17293284139523302,
      "grad_norm": 0.9509322643280029,
      "learning_rate": 4.8883829179441884e-05,
      "loss": 0.2252,
      "num_input_tokens_seen": 6157248,
      "step": 3820
    },
    {
      "epoch": 0.173159193281876,
      "grad_norm": 0.5047647953033447,
      "learning_rate": 4.888092660444682e-05,
      "loss": 0.2025,
      "num_input_tokens_seen": 6165504,
      "step": 3825
    },
    {
      "epoch": 0.17338554516851898,
      "grad_norm": 0.5509346723556519,
      "learning_rate": 4.887802034671276e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 6172960,
      "step": 3830
    },
    {
      "epoch": 0.17361189705516195,
      "grad_norm": 0.7803612947463989,
      "learning_rate": 4.88751104066879e-05,
      "loss": 0.1854,
      "num_input_tokens_seen": 6181056,
      "step": 3835
    },
    {
      "epoch": 0.17383824894180494,
      "grad_norm": 0.35519465804100037,
      "learning_rate": 4.887219678482098e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 6188864,
      "step": 3840
    },
    {
      "epoch": 0.1740646008284479,
      "grad_norm": 0.5310024619102478,
      "learning_rate": 4.8869279481561316e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 6196896,
      "step": 3845
    },
    {
      "epoch": 0.17429095271509087,
      "grad_norm": 0.4875757694244385,
      "learning_rate": 4.88663584973588e-05,
      "loss": 0.2358,
      "num_input_tokens_seen": 6205408,
      "step": 3850
    },
    {
      "epoch": 0.17451730460173387,
      "grad_norm": 0.6138989925384521,
      "learning_rate": 4.8863433832663874e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 6213536,
      "step": 3855
    },
    {
      "epoch": 0.17474365648837684,
      "grad_norm": 0.7359387278556824,
      "learning_rate": 4.886050548792757e-05,
      "loss": 0.207,
      "num_input_tokens_seen": 6221344,
      "step": 3860
    },
    {
      "epoch": 0.1749700083750198,
      "grad_norm": 0.793906033039093,
      "learning_rate": 4.8857573463601465e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 6229376,
      "step": 3865
    },
    {
      "epoch": 0.17519636026166277,
      "grad_norm": 1.123663067817688,
      "learning_rate": 4.885463776013772e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 6237536,
      "step": 3870
    },
    {
      "epoch": 0.17542271214830576,
      "grad_norm": 0.8594347238540649,
      "learning_rate": 4.8851698377989056e-05,
      "loss": 0.1927,
      "num_input_tokens_seen": 6245664,
      "step": 3875
    },
    {
      "epoch": 0.17564906403494873,
      "grad_norm": 0.5347675085067749,
      "learning_rate": 4.884875531760876e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 6253376,
      "step": 3880
    },
    {
      "epoch": 0.1758754159215917,
      "grad_norm": 0.3734123408794403,
      "learning_rate": 4.88458085794507e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 6261088,
      "step": 3885
    },
    {
      "epoch": 0.1761017678082347,
      "grad_norm": 0.8886108994483948,
      "learning_rate": 4.884285816396929e-05,
      "loss": 0.2316,
      "num_input_tokens_seen": 6269408,
      "step": 3890
    },
    {
      "epoch": 0.17632811969487766,
      "grad_norm": 0.7523662447929382,
      "learning_rate": 4.8839904071619526e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 6277664,
      "step": 3895
    },
    {
      "epoch": 0.17655447158152063,
      "grad_norm": 0.47163379192352295,
      "learning_rate": 4.8836946302856955e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 6285152,
      "step": 3900
    },
    {
      "epoch": 0.1767808234681636,
      "grad_norm": 1.1959114074707031,
      "learning_rate": 4.8833984858137715e-05,
      "loss": 0.1929,
      "num_input_tokens_seen": 6293280,
      "step": 3905
    },
    {
      "epoch": 0.1770071753548066,
      "grad_norm": 0.46329885721206665,
      "learning_rate": 4.8831019737918494e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 6301728,
      "step": 3910
    },
    {
      "epoch": 0.17723352724144956,
      "grad_norm": 0.395020067691803,
      "learning_rate": 4.882805094265655e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 6310272,
      "step": 3915
    },
    {
      "epoch": 0.17745987912809252,
      "grad_norm": 0.4688641428947449,
      "learning_rate": 4.8825078472809706e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 6317600,
      "step": 3920
    },
    {
      "epoch": 0.17768623101473552,
      "grad_norm": 1.226278305053711,
      "learning_rate": 4.882210232883635e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 6326048,
      "step": 3925
    },
    {
      "epoch": 0.17791258290137849,
      "grad_norm": 0.8641949892044067,
      "learning_rate": 4.881912251119546e-05,
      "loss": 0.2297,
      "num_input_tokens_seen": 6334016,
      "step": 3930
    },
    {
      "epoch": 0.17813893478802145,
      "grad_norm": 1.1583906412124634,
      "learning_rate": 4.881613902034654e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 6342592,
      "step": 3935
    },
    {
      "epoch": 0.17836528667466445,
      "grad_norm": 0.854123592376709,
      "learning_rate": 4.88131518567497e-05,
      "loss": 0.1967,
      "num_input_tokens_seen": 6350432,
      "step": 3940
    },
    {
      "epoch": 0.17859163856130741,
      "grad_norm": 0.47776100039482117,
      "learning_rate": 4.881016102086558e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 6358688,
      "step": 3945
    },
    {
      "epoch": 0.17881799044795038,
      "grad_norm": 0.8212628364562988,
      "learning_rate": 4.8807166513155425e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 6367360,
      "step": 3950
    },
    {
      "epoch": 0.17904434233459335,
      "grad_norm": 0.4085969924926758,
      "learning_rate": 4.8804168334081004e-05,
      "loss": 0.1742,
      "num_input_tokens_seen": 6375200,
      "step": 3955
    },
    {
      "epoch": 0.17927069422123634,
      "grad_norm": 0.43182462453842163,
      "learning_rate": 4.880116648410468e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 6383008,
      "step": 3960
    },
    {
      "epoch": 0.1794970461078793,
      "grad_norm": 1.1997958421707153,
      "learning_rate": 4.879816096368939e-05,
      "loss": 0.2037,
      "num_input_tokens_seen": 6390624,
      "step": 3965
    },
    {
      "epoch": 0.17972339799452228,
      "grad_norm": 0.8741810321807861,
      "learning_rate": 4.879515177329861e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 6398432,
      "step": 3970
    },
    {
      "epoch": 0.17994974988116527,
      "grad_norm": 0.46303805708885193,
      "learning_rate": 4.8792138913396394e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 6406688,
      "step": 3975
    },
    {
      "epoch": 0.18017610176780824,
      "grad_norm": 1.1116161346435547,
      "learning_rate": 4.8789122384447374e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 6414304,
      "step": 3980
    },
    {
      "epoch": 0.1804024536544512,
      "grad_norm": 0.27651792764663696,
      "learning_rate": 4.878610218691673e-05,
      "loss": 0.2274,
      "num_input_tokens_seen": 6422592,
      "step": 3985
    },
    {
      "epoch": 0.18062880554109417,
      "grad_norm": 0.49702557921409607,
      "learning_rate": 4.87830783212702e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 6430656,
      "step": 3990
    },
    {
      "epoch": 0.18085515742773717,
      "grad_norm": 0.6005845069885254,
      "learning_rate": 4.878005078797413e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 6439488,
      "step": 3995
    },
    {
      "epoch": 0.18108150931438013,
      "grad_norm": 0.46940377354621887,
      "learning_rate": 4.877701958749539e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 6448128,
      "step": 4000
    },
    {
      "epoch": 0.18108150931438013,
      "eval_loss": 0.18440458178520203,
      "eval_runtime": 404.3194,
      "eval_samples_per_second": 97.129,
      "eval_steps_per_second": 24.283,
      "num_input_tokens_seen": 6448128,
      "step": 4000
    },
    {
      "epoch": 0.1813078612010231,
      "grad_norm": 0.5532128810882568,
      "learning_rate": 4.877398472030142e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 6456608,
      "step": 4005
    },
    {
      "epoch": 0.1815342130876661,
      "grad_norm": 0.4724082052707672,
      "learning_rate": 4.877094618686024e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 6464864,
      "step": 4010
    },
    {
      "epoch": 0.18176056497430906,
      "grad_norm": 0.5952568054199219,
      "learning_rate": 4.876790398764045e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 6472896,
      "step": 4015
    },
    {
      "epoch": 0.18198691686095203,
      "grad_norm": 0.4474090039730072,
      "learning_rate": 4.8764858123111167e-05,
      "loss": 0.1876,
      "num_input_tokens_seen": 6481024,
      "step": 4020
    },
    {
      "epoch": 0.182213268747595,
      "grad_norm": 0.42300450801849365,
      "learning_rate": 4.876180859374212e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 6488768,
      "step": 4025
    },
    {
      "epoch": 0.182439620634238,
      "grad_norm": 0.497814416885376,
      "learning_rate": 4.875875540000357e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 6496896,
      "step": 4030
    },
    {
      "epoch": 0.18266597252088096,
      "grad_norm": 0.4261600971221924,
      "learning_rate": 4.8755698542366376e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 6505504,
      "step": 4035
    },
    {
      "epoch": 0.18289232440752393,
      "grad_norm": 0.8548309206962585,
      "learning_rate": 4.875263802130193e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 6513696,
      "step": 4040
    },
    {
      "epoch": 0.18311867629416692,
      "grad_norm": 0.7306091785430908,
      "learning_rate": 4.8749573837282207e-05,
      "loss": 0.209,
      "num_input_tokens_seen": 6521376,
      "step": 4045
    },
    {
      "epoch": 0.1833450281808099,
      "grad_norm": 0.6903974413871765,
      "learning_rate": 4.874650599077974e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 6528896,
      "step": 4050
    },
    {
      "epoch": 0.18357138006745286,
      "grad_norm": 0.3878736197948456,
      "learning_rate": 4.874343448226764e-05,
      "loss": 0.1981,
      "num_input_tokens_seen": 6537088,
      "step": 4055
    },
    {
      "epoch": 0.18379773195409585,
      "grad_norm": 0.6954119205474854,
      "learning_rate": 4.874035931221955e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 6545088,
      "step": 4060
    },
    {
      "epoch": 0.18402408384073882,
      "grad_norm": 0.8500886559486389,
      "learning_rate": 4.8737280481109724e-05,
      "loss": 0.1919,
      "num_input_tokens_seen": 6553472,
      "step": 4065
    },
    {
      "epoch": 0.18425043572738178,
      "grad_norm": 0.6793984174728394,
      "learning_rate": 4.873419798941294e-05,
      "loss": 0.2017,
      "num_input_tokens_seen": 6561568,
      "step": 4070
    },
    {
      "epoch": 0.18447678761402475,
      "grad_norm": 0.8033963441848755,
      "learning_rate": 4.873111183760458e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 6569952,
      "step": 4075
    },
    {
      "epoch": 0.18470313950066775,
      "grad_norm": 1.0596202611923218,
      "learning_rate": 4.8728022026160537e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 6577888,
      "step": 4080
    },
    {
      "epoch": 0.1849294913873107,
      "grad_norm": 0.5555120706558228,
      "learning_rate": 4.872492855555732e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 6585888,
      "step": 4085
    },
    {
      "epoch": 0.18515584327395368,
      "grad_norm": 0.4291716516017914,
      "learning_rate": 4.8721831426271956e-05,
      "loss": 0.2071,
      "num_input_tokens_seen": 6593152,
      "step": 4090
    },
    {
      "epoch": 0.18538219516059667,
      "grad_norm": 0.6672114133834839,
      "learning_rate": 4.87187306387821e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 6601056,
      "step": 4095
    },
    {
      "epoch": 0.18560854704723964,
      "grad_norm": 1.115906834602356,
      "learning_rate": 4.87156261935659e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 6608416,
      "step": 4100
    },
    {
      "epoch": 0.1858348989338826,
      "grad_norm": 0.36832377314567566,
      "learning_rate": 4.871251809110211e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 6617216,
      "step": 4105
    },
    {
      "epoch": 0.18606125082052558,
      "grad_norm": 0.5552628040313721,
      "learning_rate": 4.8709406331870044e-05,
      "loss": 0.1844,
      "num_input_tokens_seen": 6624704,
      "step": 4110
    },
    {
      "epoch": 0.18628760270716857,
      "grad_norm": 0.39153793454170227,
      "learning_rate": 4.8706290916349574e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 6632928,
      "step": 4115
    },
    {
      "epoch": 0.18651395459381154,
      "grad_norm": 0.8011826872825623,
      "learning_rate": 4.8703171845021134e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 6641344,
      "step": 4120
    },
    {
      "epoch": 0.1867403064804545,
      "grad_norm": 1.0655124187469482,
      "learning_rate": 4.870004911836572e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 6648960,
      "step": 4125
    },
    {
      "epoch": 0.1869666583670975,
      "grad_norm": 0.7629068493843079,
      "learning_rate": 4.869692273686489e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 6657280,
      "step": 4130
    },
    {
      "epoch": 0.18719301025374047,
      "grad_norm": 0.872664213180542,
      "learning_rate": 4.869379270100079e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 6665568,
      "step": 4135
    },
    {
      "epoch": 0.18741936214038343,
      "grad_norm": 0.5162887573242188,
      "learning_rate": 4.86906590112561e-05,
      "loss": 0.2081,
      "num_input_tokens_seen": 6673504,
      "step": 4140
    },
    {
      "epoch": 0.18764571402702643,
      "grad_norm": 0.8384267091751099,
      "learning_rate": 4.8687521668114064e-05,
      "loss": 0.2032,
      "num_input_tokens_seen": 6681664,
      "step": 4145
    },
    {
      "epoch": 0.1878720659136694,
      "grad_norm": 0.4885372221469879,
      "learning_rate": 4.868438067205853e-05,
      "loss": 0.1907,
      "num_input_tokens_seen": 6690144,
      "step": 4150
    },
    {
      "epoch": 0.18809841780031236,
      "grad_norm": 0.40804946422576904,
      "learning_rate": 4.8681236023573844e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 6698144,
      "step": 4155
    },
    {
      "epoch": 0.18832476968695533,
      "grad_norm": 0.5952759385108948,
      "learning_rate": 4.867808772314497e-05,
      "loss": 0.213,
      "num_input_tokens_seen": 6706880,
      "step": 4160
    },
    {
      "epoch": 0.18855112157359832,
      "grad_norm": 0.35358014702796936,
      "learning_rate": 4.867493577125741e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 6714880,
      "step": 4165
    },
    {
      "epoch": 0.1887774734602413,
      "grad_norm": 0.7194891571998596,
      "learning_rate": 4.867178016839725e-05,
      "loss": 0.1898,
      "num_input_tokens_seen": 6722368,
      "step": 4170
    },
    {
      "epoch": 0.18900382534688426,
      "grad_norm": 0.5760931968688965,
      "learning_rate": 4.8668620915051094e-05,
      "loss": 0.1919,
      "num_input_tokens_seen": 6730656,
      "step": 4175
    },
    {
      "epoch": 0.18923017723352725,
      "grad_norm": 0.5877888798713684,
      "learning_rate": 4.866545801170616e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 6738816,
      "step": 4180
    },
    {
      "epoch": 0.18945652912017022,
      "grad_norm": 0.5256401896476746,
      "learning_rate": 4.86622914588502e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 6747520,
      "step": 4185
    },
    {
      "epoch": 0.1896828810068132,
      "grad_norm": 0.7563493847846985,
      "learning_rate": 4.865912125697154e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 6755648,
      "step": 4190
    },
    {
      "epoch": 0.18990923289345615,
      "grad_norm": 0.5360850691795349,
      "learning_rate": 4.865594740655907e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 6763104,
      "step": 4195
    },
    {
      "epoch": 0.19013558478009915,
      "grad_norm": 0.6639833450317383,
      "learning_rate": 4.865276990810222e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 6771360,
      "step": 4200
    },
    {
      "epoch": 0.19013558478009915,
      "eval_loss": 0.18300482630729675,
      "eval_runtime": 404.3344,
      "eval_samples_per_second": 97.125,
      "eval_steps_per_second": 24.282,
      "num_input_tokens_seen": 6771360,
      "step": 4200
    },
    {
      "epoch": 0.19036193666674212,
      "grad_norm": 1.325181245803833,
      "learning_rate": 4.8649588762091016e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 6779488,
      "step": 4205
    },
    {
      "epoch": 0.19058828855338508,
      "grad_norm": 0.22959744930267334,
      "learning_rate": 4.8646403969016016e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 6786784,
      "step": 4210
    },
    {
      "epoch": 0.19081464044002808,
      "grad_norm": 0.6897128820419312,
      "learning_rate": 4.864321552936838e-05,
      "loss": 0.2163,
      "num_input_tokens_seen": 6794560,
      "step": 4215
    },
    {
      "epoch": 0.19104099232667104,
      "grad_norm": 0.3829789459705353,
      "learning_rate": 4.864002344363978e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 6802560,
      "step": 4220
    },
    {
      "epoch": 0.191267344213314,
      "grad_norm": 1.2819255590438843,
      "learning_rate": 4.863682771232248e-05,
      "loss": 0.2368,
      "num_input_tokens_seen": 6810752,
      "step": 4225
    },
    {
      "epoch": 0.191493696099957,
      "grad_norm": 1.0932648181915283,
      "learning_rate": 4.8633628335909324e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 6818720,
      "step": 4230
    },
    {
      "epoch": 0.19172004798659997,
      "grad_norm": 0.793010950088501,
      "learning_rate": 4.8630425314893676e-05,
      "loss": 0.1881,
      "num_input_tokens_seen": 6826880,
      "step": 4235
    },
    {
      "epoch": 0.19194639987324294,
      "grad_norm": 0.4907114505767822,
      "learning_rate": 4.862721864976948e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 6834784,
      "step": 4240
    },
    {
      "epoch": 0.1921727517598859,
      "grad_norm": 0.6480650305747986,
      "learning_rate": 4.862400834103125e-05,
      "loss": 0.1913,
      "num_input_tokens_seen": 6842784,
      "step": 4245
    },
    {
      "epoch": 0.1923991036465289,
      "grad_norm": 0.8681305646896362,
      "learning_rate": 4.862079438917406e-05,
      "loss": 0.2,
      "num_input_tokens_seen": 6851040,
      "step": 4250
    },
    {
      "epoch": 0.19262545553317187,
      "grad_norm": 0.5637214183807373,
      "learning_rate": 4.8617576794693536e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 6859008,
      "step": 4255
    },
    {
      "epoch": 0.19285180741981484,
      "grad_norm": 0.7615519165992737,
      "learning_rate": 4.8614355558085875e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 6866976,
      "step": 4260
    },
    {
      "epoch": 0.19307815930645783,
      "grad_norm": 0.40713414549827576,
      "learning_rate": 4.861113067984783e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 6874976,
      "step": 4265
    },
    {
      "epoch": 0.1933045111931008,
      "grad_norm": 0.934788703918457,
      "learning_rate": 4.860790216047671e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 6883680,
      "step": 4270
    },
    {
      "epoch": 0.19353086307974376,
      "grad_norm": 1.1625192165374756,
      "learning_rate": 4.860467000047041e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 6892032,
      "step": 4275
    },
    {
      "epoch": 0.19375721496638673,
      "grad_norm": 0.4595896303653717,
      "learning_rate": 4.860143420032737e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 6899712,
      "step": 4280
    },
    {
      "epoch": 0.19398356685302973,
      "grad_norm": 0.912107527256012,
      "learning_rate": 4.859819476054657e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 6907744,
      "step": 4285
    },
    {
      "epoch": 0.1942099187396727,
      "grad_norm": 0.5052017569541931,
      "learning_rate": 4.859495168162758e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 6915968,
      "step": 4290
    },
    {
      "epoch": 0.19443627062631566,
      "grad_norm": 1.0234121084213257,
      "learning_rate": 4.859170496407054e-05,
      "loss": 0.1916,
      "num_input_tokens_seen": 6924096,
      "step": 4295
    },
    {
      "epoch": 0.19466262251295866,
      "grad_norm": 0.9395620822906494,
      "learning_rate": 4.8588454608376114e-05,
      "loss": 0.2025,
      "num_input_tokens_seen": 6932000,
      "step": 4300
    },
    {
      "epoch": 0.19488897439960162,
      "grad_norm": 0.49298375844955444,
      "learning_rate": 4.8585200615045555e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 6939904,
      "step": 4305
    },
    {
      "epoch": 0.1951153262862446,
      "grad_norm": 0.5162792801856995,
      "learning_rate": 4.8581942984580674e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 6948032,
      "step": 4310
    },
    {
      "epoch": 0.19534167817288758,
      "grad_norm": 0.7268537878990173,
      "learning_rate": 4.857868171748384e-05,
      "loss": 0.1709,
      "num_input_tokens_seen": 6955968,
      "step": 4315
    },
    {
      "epoch": 0.19556803005953055,
      "grad_norm": 0.7250018119812012,
      "learning_rate": 4.8575416814257976e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 6963712,
      "step": 4320
    },
    {
      "epoch": 0.19579438194617352,
      "grad_norm": 0.5118347406387329,
      "learning_rate": 4.857214827540657e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 6971904,
      "step": 4325
    },
    {
      "epoch": 0.19602073383281648,
      "grad_norm": 1.1366809606552124,
      "learning_rate": 4.856887610143367e-05,
      "loss": 0.1981,
      "num_input_tokens_seen": 6980288,
      "step": 4330
    },
    {
      "epoch": 0.19624708571945948,
      "grad_norm": 1.3405237197875977,
      "learning_rate": 4.8565600292843896e-05,
      "loss": 0.2066,
      "num_input_tokens_seen": 6988480,
      "step": 4335
    },
    {
      "epoch": 0.19647343760610245,
      "grad_norm": 0.5073785185813904,
      "learning_rate": 4.856232085014241e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 6996608,
      "step": 4340
    },
    {
      "epoch": 0.1966997894927454,
      "grad_norm": 0.25917232036590576,
      "learning_rate": 4.855903777383495e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 7004256,
      "step": 4345
    },
    {
      "epoch": 0.1969261413793884,
      "grad_norm": 0.9761782884597778,
      "learning_rate": 4.85557510644278e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 7012224,
      "step": 4350
    },
    {
      "epoch": 0.19715249326603138,
      "grad_norm": 0.5834032297134399,
      "learning_rate": 4.855246072242782e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 7020992,
      "step": 4355
    },
    {
      "epoch": 0.19737884515267434,
      "grad_norm": 1.039244532585144,
      "learning_rate": 4.8549166748342414e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 7028960,
      "step": 4360
    },
    {
      "epoch": 0.1976051970393173,
      "grad_norm": 0.6926798820495605,
      "learning_rate": 4.8545869142679556e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 7036992,
      "step": 4365
    },
    {
      "epoch": 0.1978315489259603,
      "grad_norm": 1.3988780975341797,
      "learning_rate": 4.8542567905947776e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 7047776,
      "step": 4370
    },
    {
      "epoch": 0.19805790081260327,
      "grad_norm": 0.4820401668548584,
      "learning_rate": 4.853926303865618e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 7055968,
      "step": 4375
    },
    {
      "epoch": 0.19828425269924624,
      "grad_norm": 0.8160607814788818,
      "learning_rate": 4.853595454131441e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 7064416,
      "step": 4380
    },
    {
      "epoch": 0.19851060458588923,
      "grad_norm": 0.46490028500556946,
      "learning_rate": 4.8532642414432674e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 7072192,
      "step": 4385
    },
    {
      "epoch": 0.1987369564725322,
      "grad_norm": 0.6240481734275818,
      "learning_rate": 4.8529326658521754e-05,
      "loss": 0.1862,
      "num_input_tokens_seen": 7080128,
      "step": 4390
    },
    {
      "epoch": 0.19896330835917517,
      "grad_norm": 0.5689383149147034,
      "learning_rate": 4.8526007274092965e-05,
      "loss": 0.2001,
      "num_input_tokens_seen": 7088736,
      "step": 4395
    },
    {
      "epoch": 0.19918966024581816,
      "grad_norm": 0.7058113217353821,
      "learning_rate": 4.852268426165822e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 7097088,
      "step": 4400
    },
    {
      "epoch": 0.19918966024581816,
      "eval_loss": 0.18204113841056824,
      "eval_runtime": 404.7479,
      "eval_samples_per_second": 97.026,
      "eval_steps_per_second": 24.257,
      "num_input_tokens_seen": 7097088,
      "step": 4400
    },
    {
      "epoch": 0.19941601213246113,
      "grad_norm": 0.9841000437736511,
      "learning_rate": 4.851935762172995e-05,
      "loss": 0.1987,
      "num_input_tokens_seen": 7105056,
      "step": 4405
    },
    {
      "epoch": 0.1996423640191041,
      "grad_norm": 0.4904654920101166,
      "learning_rate": 4.8516027354821175e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 7113312,
      "step": 4410
    },
    {
      "epoch": 0.19986871590574706,
      "grad_norm": 0.3379433751106262,
      "learning_rate": 4.851269346144546e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 7121280,
      "step": 4415
    },
    {
      "epoch": 0.20009506779239006,
      "grad_norm": 0.5913249254226685,
      "learning_rate": 4.850935594211693e-05,
      "loss": 0.2209,
      "num_input_tokens_seen": 7129376,
      "step": 4420
    },
    {
      "epoch": 0.20032141967903302,
      "grad_norm": 0.5537619590759277,
      "learning_rate": 4.850601479735029e-05,
      "loss": 0.1746,
      "num_input_tokens_seen": 7137504,
      "step": 4425
    },
    {
      "epoch": 0.200547771565676,
      "grad_norm": 0.5040625333786011,
      "learning_rate": 4.850267002766076e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 7145408,
      "step": 4430
    },
    {
      "epoch": 0.200774123452319,
      "grad_norm": 0.6897289752960205,
      "learning_rate": 4.849932163356417e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 7153984,
      "step": 4435
    },
    {
      "epoch": 0.20100047533896195,
      "grad_norm": 0.9050817489624023,
      "learning_rate": 4.8495969615576864e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 7161344,
      "step": 4440
    },
    {
      "epoch": 0.20122682722560492,
      "grad_norm": 0.6501354575157166,
      "learning_rate": 4.849261397421577e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 7169248,
      "step": 4445
    },
    {
      "epoch": 0.2014531791122479,
      "grad_norm": 0.4095616638660431,
      "learning_rate": 4.848925470999839e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 7176960,
      "step": 4450
    },
    {
      "epoch": 0.20167953099889088,
      "grad_norm": 0.5594167709350586,
      "learning_rate": 4.848589182344273e-05,
      "loss": 0.1824,
      "num_input_tokens_seen": 7184928,
      "step": 4455
    },
    {
      "epoch": 0.20190588288553385,
      "grad_norm": 0.5390444397926331,
      "learning_rate": 4.848252531506742e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 7193184,
      "step": 4460
    },
    {
      "epoch": 0.20213223477217682,
      "grad_norm": 0.7107042074203491,
      "learning_rate": 4.847915518539161e-05,
      "loss": 0.1746,
      "num_input_tokens_seen": 7201664,
      "step": 4465
    },
    {
      "epoch": 0.2023585866588198,
      "grad_norm": 0.3377852439880371,
      "learning_rate": 4.847578143493501e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 7209888,
      "step": 4470
    },
    {
      "epoch": 0.20258493854546278,
      "grad_norm": 0.5312425494194031,
      "learning_rate": 4.847240406421789e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 7217472,
      "step": 4475
    },
    {
      "epoch": 0.20281129043210575,
      "grad_norm": 0.5856176614761353,
      "learning_rate": 4.84690230737611e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 7225376,
      "step": 4480
    },
    {
      "epoch": 0.20303764231874874,
      "grad_norm": 1.098870038986206,
      "learning_rate": 4.846563846408602e-05,
      "loss": 0.1854,
      "num_input_tokens_seen": 7233216,
      "step": 4485
    },
    {
      "epoch": 0.2032639942053917,
      "grad_norm": 0.8568993806838989,
      "learning_rate": 4.84622502357146e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 7241120,
      "step": 4490
    },
    {
      "epoch": 0.20349034609203467,
      "grad_norm": 0.869484007358551,
      "learning_rate": 4.8458858389169345e-05,
      "loss": 0.2031,
      "num_input_tokens_seen": 7248992,
      "step": 4495
    },
    {
      "epoch": 0.20371669797867764,
      "grad_norm": 0.44272175431251526,
      "learning_rate": 4.8455462924973334e-05,
      "loss": 0.2043,
      "num_input_tokens_seen": 7256800,
      "step": 4500
    },
    {
      "epoch": 0.20394304986532064,
      "grad_norm": 0.6057604551315308,
      "learning_rate": 4.845206384365018e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 7265152,
      "step": 4505
    },
    {
      "epoch": 0.2041694017519636,
      "grad_norm": 0.582863450050354,
      "learning_rate": 4.844866114572405e-05,
      "loss": 0.2166,
      "num_input_tokens_seen": 7272992,
      "step": 4510
    },
    {
      "epoch": 0.20439575363860657,
      "grad_norm": 0.5405785441398621,
      "learning_rate": 4.8445254831719706e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 7280928,
      "step": 4515
    },
    {
      "epoch": 0.20462210552524956,
      "grad_norm": 0.6572803258895874,
      "learning_rate": 4.8441844902162434e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 7289056,
      "step": 4520
    },
    {
      "epoch": 0.20484845741189253,
      "grad_norm": 1.0986084938049316,
      "learning_rate": 4.843843135757809e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 7296864,
      "step": 4525
    },
    {
      "epoch": 0.2050748092985355,
      "grad_norm": 0.946144163608551,
      "learning_rate": 4.843501419849308e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 7305248,
      "step": 4530
    },
    {
      "epoch": 0.20530116118517847,
      "grad_norm": 0.6319920420646667,
      "learning_rate": 4.8431593425434386e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 7313408,
      "step": 4535
    },
    {
      "epoch": 0.20552751307182146,
      "grad_norm": 0.937760055065155,
      "learning_rate": 4.8428169038929526e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 7322464,
      "step": 4540
    },
    {
      "epoch": 0.20575386495846443,
      "grad_norm": 0.5224272012710571,
      "learning_rate": 4.8424741039506575e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 7330784,
      "step": 4545
    },
    {
      "epoch": 0.2059802168451074,
      "grad_norm": 0.3947784900665283,
      "learning_rate": 4.842130942769419e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 7339040,
      "step": 4550
    },
    {
      "epoch": 0.2062065687317504,
      "grad_norm": 0.34507647156715393,
      "learning_rate": 4.841787420402156e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 7347296,
      "step": 4555
    },
    {
      "epoch": 0.20643292061839336,
      "grad_norm": 0.5546767115592957,
      "learning_rate": 4.841443536901844e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 7355680,
      "step": 4560
    },
    {
      "epoch": 0.20665927250503632,
      "grad_norm": 0.4155840277671814,
      "learning_rate": 4.841099292321514e-05,
      "loss": 0.2228,
      "num_input_tokens_seen": 7363168,
      "step": 4565
    },
    {
      "epoch": 0.20688562439167932,
      "grad_norm": 0.4277958273887634,
      "learning_rate": 4.8407546867142525e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 7371232,
      "step": 4570
    },
    {
      "epoch": 0.20711197627832229,
      "grad_norm": 0.6233193874359131,
      "learning_rate": 4.840409720133203e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 7379072,
      "step": 4575
    },
    {
      "epoch": 0.20733832816496525,
      "grad_norm": 0.6338982582092285,
      "learning_rate": 4.8400643926315634e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 7387200,
      "step": 4580
    },
    {
      "epoch": 0.20756468005160822,
      "grad_norm": 1.0142478942871094,
      "learning_rate": 4.839718704262587e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 7394944,
      "step": 4585
    },
    {
      "epoch": 0.2077910319382512,
      "grad_norm": 1.006313443183899,
      "learning_rate": 4.839372655079585e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 7402944,
      "step": 4590
    },
    {
      "epoch": 0.20801738382489418,
      "grad_norm": 0.5090396404266357,
      "learning_rate": 4.83902624513592e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 7410752,
      "step": 4595
    },
    {
      "epoch": 0.20824373571153715,
      "grad_norm": 1.055311679840088,
      "learning_rate": 4.838679474485014e-05,
      "loss": 0.1514,
      "num_input_tokens_seen": 7418432,
      "step": 4600
    },
    {
      "epoch": 0.20824373571153715,
      "eval_loss": 0.18061643838882446,
      "eval_runtime": 404.7655,
      "eval_samples_per_second": 97.022,
      "eval_steps_per_second": 24.256,
      "num_input_tokens_seen": 7418432,
      "step": 4600
    },
    {
      "epoch": 0.20847008759818014,
      "grad_norm": 0.8971060514450073,
      "learning_rate": 4.838332343180343e-05,
      "loss": 0.1997,
      "num_input_tokens_seen": 7426304,
      "step": 4605
    },
    {
      "epoch": 0.2086964394848231,
      "grad_norm": 0.557102382183075,
      "learning_rate": 4.83798485127544e-05,
      "loss": 0.2143,
      "num_input_tokens_seen": 7434336,
      "step": 4610
    },
    {
      "epoch": 0.20892279137146608,
      "grad_norm": 0.5773736834526062,
      "learning_rate": 4.837636998823892e-05,
      "loss": 0.1787,
      "num_input_tokens_seen": 7442272,
      "step": 4615
    },
    {
      "epoch": 0.20914914325810904,
      "grad_norm": 0.4418168067932129,
      "learning_rate": 4.8372887858793414e-05,
      "loss": 0.2029,
      "num_input_tokens_seen": 7449920,
      "step": 4620
    },
    {
      "epoch": 0.20937549514475204,
      "grad_norm": 0.317018061876297,
      "learning_rate": 4.836940212495489e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 7457536,
      "step": 4625
    },
    {
      "epoch": 0.209601847031395,
      "grad_norm": 0.477145254611969,
      "learning_rate": 4.836591278726087e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 7465728,
      "step": 4630
    },
    {
      "epoch": 0.20982819891803797,
      "grad_norm": 0.42327651381492615,
      "learning_rate": 4.836241984624947e-05,
      "loss": 0.2117,
      "num_input_tokens_seen": 7473312,
      "step": 4635
    },
    {
      "epoch": 0.21005455080468097,
      "grad_norm": 0.6948663592338562,
      "learning_rate": 4.8358923302459336e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 7480704,
      "step": 4640
    },
    {
      "epoch": 0.21028090269132393,
      "grad_norm": 0.3899434506893158,
      "learning_rate": 4.835542315642968e-05,
      "loss": 0.2151,
      "num_input_tokens_seen": 7489344,
      "step": 4645
    },
    {
      "epoch": 0.2105072545779669,
      "grad_norm": 0.4928063154220581,
      "learning_rate": 4.8351919408700274e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 7497088,
      "step": 4650
    },
    {
      "epoch": 0.2107336064646099,
      "grad_norm": 0.6403312683105469,
      "learning_rate": 4.834841205981144e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 7505088,
      "step": 4655
    },
    {
      "epoch": 0.21095995835125286,
      "grad_norm": 0.5767378807067871,
      "learning_rate": 4.8344901110304054e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 7513216,
      "step": 4660
    },
    {
      "epoch": 0.21118631023789583,
      "grad_norm": 1.0729047060012817,
      "learning_rate": 4.8341386560719534e-05,
      "loss": 0.2207,
      "num_input_tokens_seen": 7520864,
      "step": 4665
    },
    {
      "epoch": 0.2114126621245388,
      "grad_norm": 0.4617128372192383,
      "learning_rate": 4.833786841159989e-05,
      "loss": 0.1992,
      "num_input_tokens_seen": 7529664,
      "step": 4670
    },
    {
      "epoch": 0.2116390140111818,
      "grad_norm": 0.6750272512435913,
      "learning_rate": 4.833434666348765e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 7538016,
      "step": 4675
    },
    {
      "epoch": 0.21186536589782476,
      "grad_norm": 0.3500758409500122,
      "learning_rate": 4.833082131692592e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 7546240,
      "step": 4680
    },
    {
      "epoch": 0.21209171778446773,
      "grad_norm": 0.468941867351532,
      "learning_rate": 4.832729237245835e-05,
      "loss": 0.2383,
      "num_input_tokens_seen": 7554336,
      "step": 4685
    },
    {
      "epoch": 0.21231806967111072,
      "grad_norm": 0.4065745174884796,
      "learning_rate": 4.8323759830629145e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 7561984,
      "step": 4690
    },
    {
      "epoch": 0.2125444215577537,
      "grad_norm": 0.5818890929222107,
      "learning_rate": 4.8320223691983066e-05,
      "loss": 0.1827,
      "num_input_tokens_seen": 7569920,
      "step": 4695
    },
    {
      "epoch": 0.21277077344439665,
      "grad_norm": 0.29089945554733276,
      "learning_rate": 4.831668395706544e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 7577824,
      "step": 4700
    },
    {
      "epoch": 0.21299712533103962,
      "grad_norm": 0.7983177304267883,
      "learning_rate": 4.8313140626422125e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 7585696,
      "step": 4705
    },
    {
      "epoch": 0.21322347721768262,
      "grad_norm": 0.3724236488342285,
      "learning_rate": 4.830959370059956e-05,
      "loss": 0.2085,
      "num_input_tokens_seen": 7593920,
      "step": 4710
    },
    {
      "epoch": 0.21344982910432558,
      "grad_norm": 0.8408177495002747,
      "learning_rate": 4.830604318014472e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 7602112,
      "step": 4715
    },
    {
      "epoch": 0.21367618099096855,
      "grad_norm": 0.544820249080658,
      "learning_rate": 4.830248906560514e-05,
      "loss": 0.1837,
      "num_input_tokens_seen": 7610624,
      "step": 4720
    },
    {
      "epoch": 0.21390253287761155,
      "grad_norm": 0.2766430079936981,
      "learning_rate": 4.829893135752891e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 7618336,
      "step": 4725
    },
    {
      "epoch": 0.2141288847642545,
      "grad_norm": 0.7350754737854004,
      "learning_rate": 4.829537005646466e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 7626272,
      "step": 4730
    },
    {
      "epoch": 0.21435523665089748,
      "grad_norm": 0.3916795551776886,
      "learning_rate": 4.8291805162961615e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 7635008,
      "step": 4735
    },
    {
      "epoch": 0.21458158853754047,
      "grad_norm": 0.6082535982131958,
      "learning_rate": 4.82882366775695e-05,
      "loss": 0.1929,
      "num_input_tokens_seen": 7643072,
      "step": 4740
    },
    {
      "epoch": 0.21480794042418344,
      "grad_norm": 0.9162431955337524,
      "learning_rate": 4.828466460083864e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 7650720,
      "step": 4745
    },
    {
      "epoch": 0.2150342923108264,
      "grad_norm": 0.30432823300361633,
      "learning_rate": 4.8281088933319877e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 7658720,
      "step": 4750
    },
    {
      "epoch": 0.21526064419746938,
      "grad_norm": 0.7134421467781067,
      "learning_rate": 4.827750967556464e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 7666176,
      "step": 4755
    },
    {
      "epoch": 0.21548699608411237,
      "grad_norm": 0.6022066473960876,
      "learning_rate": 4.827392682812488e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 7674784,
      "step": 4760
    },
    {
      "epoch": 0.21571334797075534,
      "grad_norm": 0.7610297799110413,
      "learning_rate": 4.827034039155312e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 7683840,
      "step": 4765
    },
    {
      "epoch": 0.2159396998573983,
      "grad_norm": 1.0409704446792603,
      "learning_rate": 4.8266750366402445e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 7691552,
      "step": 4770
    },
    {
      "epoch": 0.2161660517440413,
      "grad_norm": 0.5172185301780701,
      "learning_rate": 4.8263156753226476e-05,
      "loss": 0.1954,
      "num_input_tokens_seen": 7699584,
      "step": 4775
    },
    {
      "epoch": 0.21639240363068427,
      "grad_norm": 0.5492791533470154,
      "learning_rate": 4.8259559552579394e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 7707168,
      "step": 4780
    },
    {
      "epoch": 0.21661875551732723,
      "grad_norm": 0.6497849225997925,
      "learning_rate": 4.825595876501593e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 7715200,
      "step": 4785
    },
    {
      "epoch": 0.2168451074039702,
      "grad_norm": 0.4864634573459625,
      "learning_rate": 4.825235439109137e-05,
      "loss": 0.1943,
      "num_input_tokens_seen": 7723488,
      "step": 4790
    },
    {
      "epoch": 0.2170714592906132,
      "grad_norm": 0.5292195677757263,
      "learning_rate": 4.824874643136156e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 7731392,
      "step": 4795
    },
    {
      "epoch": 0.21729781117725616,
      "grad_norm": 0.5774527788162231,
      "learning_rate": 4.824513488638288e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 7739200,
      "step": 4800
    },
    {
      "epoch": 0.21729781117725616,
      "eval_loss": 0.17822803556919098,
      "eval_runtime": 404.4337,
      "eval_samples_per_second": 97.101,
      "eval_steps_per_second": 24.276,
      "num_input_tokens_seen": 7739200,
      "step": 4800
    },
    {
      "epoch": 0.21752416306389913,
      "grad_norm": 0.5118679404258728,
      "learning_rate": 4.8241519756712293e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 7746720,
      "step": 4805
    },
    {
      "epoch": 0.21775051495054212,
      "grad_norm": 0.7038029432296753,
      "learning_rate": 4.8237901042907285e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 7754944,
      "step": 4810
    },
    {
      "epoch": 0.2179768668371851,
      "grad_norm": 0.54521244764328,
      "learning_rate": 4.823427874552591e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 7763328,
      "step": 4815
    },
    {
      "epoch": 0.21820321872382806,
      "grad_norm": 0.4025432765483856,
      "learning_rate": 4.823065286512677e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 7771424,
      "step": 4820
    },
    {
      "epoch": 0.21842957061047102,
      "grad_norm": 0.7260103821754456,
      "learning_rate": 4.8227023402269025e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 7779200,
      "step": 4825
    },
    {
      "epoch": 0.21865592249711402,
      "grad_norm": 0.9434500932693481,
      "learning_rate": 4.822339035751239e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 7787264,
      "step": 4830
    },
    {
      "epoch": 0.218882274383757,
      "grad_norm": 0.5620076060295105,
      "learning_rate": 4.8219753731417104e-05,
      "loss": 0.2122,
      "num_input_tokens_seen": 7795424,
      "step": 4835
    },
    {
      "epoch": 0.21910862627039995,
      "grad_norm": 0.6521144509315491,
      "learning_rate": 4.821611352454401e-05,
      "loss": 0.1785,
      "num_input_tokens_seen": 7803072,
      "step": 4840
    },
    {
      "epoch": 0.21933497815704295,
      "grad_norm": 0.4602662920951843,
      "learning_rate": 4.8212469737454444e-05,
      "loss": 0.1954,
      "num_input_tokens_seen": 7811328,
      "step": 4845
    },
    {
      "epoch": 0.21956133004368591,
      "grad_norm": 1.5454940795898438,
      "learning_rate": 4.820882237071035e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 7819648,
      "step": 4850
    },
    {
      "epoch": 0.21978768193032888,
      "grad_norm": 0.4636555016040802,
      "learning_rate": 4.820517142487417e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 7827616,
      "step": 4855
    },
    {
      "epoch": 0.22001403381697188,
      "grad_norm": 1.0984888076782227,
      "learning_rate": 4.8201516900508956e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 7835552,
      "step": 4860
    },
    {
      "epoch": 0.22024038570361484,
      "grad_norm": 0.33623117208480835,
      "learning_rate": 4.819785879817827e-05,
      "loss": 0.206,
      "num_input_tokens_seen": 7843424,
      "step": 4865
    },
    {
      "epoch": 0.2204667375902578,
      "grad_norm": 1.0831787586212158,
      "learning_rate": 4.8194197118446226e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 7851648,
      "step": 4870
    },
    {
      "epoch": 0.22069308947690078,
      "grad_norm": 0.6163662672042847,
      "learning_rate": 4.819053186187752e-05,
      "loss": 0.2394,
      "num_input_tokens_seen": 7859392,
      "step": 4875
    },
    {
      "epoch": 0.22091944136354377,
      "grad_norm": 0.814310610294342,
      "learning_rate": 4.818686302903736e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 7867584,
      "step": 4880
    },
    {
      "epoch": 0.22114579325018674,
      "grad_norm": 0.9776266813278198,
      "learning_rate": 4.818319062049154e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 7876160,
      "step": 4885
    },
    {
      "epoch": 0.2213721451368297,
      "grad_norm": 0.6952139139175415,
      "learning_rate": 4.817951463680639e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 7885216,
      "step": 4890
    },
    {
      "epoch": 0.2215984970234727,
      "grad_norm": 0.4908582270145416,
      "learning_rate": 4.817583507854879e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 7893504,
      "step": 4895
    },
    {
      "epoch": 0.22182484891011567,
      "grad_norm": 0.49509263038635254,
      "learning_rate": 4.817215194628617e-05,
      "loss": 0.2032,
      "num_input_tokens_seen": 7900992,
      "step": 4900
    },
    {
      "epoch": 0.22205120079675864,
      "grad_norm": 0.5082409381866455,
      "learning_rate": 4.816846524058653e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 7908640,
      "step": 4905
    },
    {
      "epoch": 0.2222775526834016,
      "grad_norm": 0.35557329654693604,
      "learning_rate": 4.816477496201839e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 7916576,
      "step": 4910
    },
    {
      "epoch": 0.2225039045700446,
      "grad_norm": 0.7174779176712036,
      "learning_rate": 4.8161081111150845e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 7924416,
      "step": 4915
    },
    {
      "epoch": 0.22273025645668756,
      "grad_norm": 1.803898811340332,
      "learning_rate": 4.815738368855354e-05,
      "loss": 0.2328,
      "num_input_tokens_seen": 7932032,
      "step": 4920
    },
    {
      "epoch": 0.22295660834333053,
      "grad_norm": 1.1438313722610474,
      "learning_rate": 4.815368269479664e-05,
      "loss": 0.2345,
      "num_input_tokens_seen": 7940032,
      "step": 4925
    },
    {
      "epoch": 0.22318296022997353,
      "grad_norm": 0.4023359417915344,
      "learning_rate": 4.814997813045092e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 7948576,
      "step": 4930
    },
    {
      "epoch": 0.2234093121166165,
      "grad_norm": 0.41560977697372437,
      "learning_rate": 4.814626999608764e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 7956768,
      "step": 4935
    },
    {
      "epoch": 0.22363566400325946,
      "grad_norm": 0.3417685925960541,
      "learning_rate": 4.814255829227865e-05,
      "loss": 0.1782,
      "num_input_tokens_seen": 7965056,
      "step": 4940
    },
    {
      "epoch": 0.22386201588990245,
      "grad_norm": 1.0648486614227295,
      "learning_rate": 4.813884301959635e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 7972864,
      "step": 4945
    },
    {
      "epoch": 0.22408836777654542,
      "grad_norm": 0.4957357347011566,
      "learning_rate": 4.813512417861368e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 7980928,
      "step": 4950
    },
    {
      "epoch": 0.2243147196631884,
      "grad_norm": 0.41570428013801575,
      "learning_rate": 4.813140176990411e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 7988896,
      "step": 4955
    },
    {
      "epoch": 0.22454107154983136,
      "grad_norm": 0.6592147350311279,
      "learning_rate": 4.8127675794041714e-05,
      "loss": 0.1921,
      "num_input_tokens_seen": 7996928,
      "step": 4960
    },
    {
      "epoch": 0.22476742343647435,
      "grad_norm": 0.4093853235244751,
      "learning_rate": 4.812394625160107e-05,
      "loss": 0.2129,
      "num_input_tokens_seen": 8005856,
      "step": 4965
    },
    {
      "epoch": 0.22499377532311732,
      "grad_norm": 0.3765868842601776,
      "learning_rate": 4.812021314315732e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 8013536,
      "step": 4970
    },
    {
      "epoch": 0.22522012720976028,
      "grad_norm": 1.1249700784683228,
      "learning_rate": 4.811647646928616e-05,
      "loss": 0.1888,
      "num_input_tokens_seen": 8022432,
      "step": 4975
    },
    {
      "epoch": 0.22544647909640328,
      "grad_norm": 0.5105119943618774,
      "learning_rate": 4.8112736230563814e-05,
      "loss": 0.2129,
      "num_input_tokens_seen": 8030624,
      "step": 4980
    },
    {
      "epoch": 0.22567283098304625,
      "grad_norm": 0.5430009365081787,
      "learning_rate": 4.81089924275671e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 8038304,
      "step": 4985
    },
    {
      "epoch": 0.2258991828696892,
      "grad_norm": 0.37760263681411743,
      "learning_rate": 4.810524506087335e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 8045920,
      "step": 4990
    },
    {
      "epoch": 0.22612553475633218,
      "grad_norm": 0.4833400249481201,
      "learning_rate": 4.810149413106044e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 8053408,
      "step": 4995
    },
    {
      "epoch": 0.22635188664297518,
      "grad_norm": 0.7298837304115295,
      "learning_rate": 4.809773963870684e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 8062528,
      "step": 5000
    },
    {
      "epoch": 0.22635188664297518,
      "eval_loss": 0.17696167528629303,
      "eval_runtime": 404.9372,
      "eval_samples_per_second": 96.98,
      "eval_steps_per_second": 24.246,
      "num_input_tokens_seen": 8062528,
      "step": 5000
    },
    {
      "epoch": 0.22657823852961814,
      "grad_norm": 0.9372361302375793,
      "learning_rate": 4.809398158439151e-05,
      "loss": 0.2217,
      "num_input_tokens_seen": 8070112,
      "step": 5005
    },
    {
      "epoch": 0.2268045904162611,
      "grad_norm": 0.41676265001296997,
      "learning_rate": 4.8090219968694005e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 8078048,
      "step": 5010
    },
    {
      "epoch": 0.2270309423029041,
      "grad_norm": 0.9591243863105774,
      "learning_rate": 4.808645479219442e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 8086272,
      "step": 5015
    },
    {
      "epoch": 0.22725729418954707,
      "grad_norm": 0.5891559720039368,
      "learning_rate": 4.8082686055473375e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 8094208,
      "step": 5020
    },
    {
      "epoch": 0.22748364607619004,
      "grad_norm": 0.9674094319343567,
      "learning_rate": 4.8078913759112066e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 8102176,
      "step": 5025
    },
    {
      "epoch": 0.22770999796283303,
      "grad_norm": 0.8760989904403687,
      "learning_rate": 4.807513790369223e-05,
      "loss": 0.1856,
      "num_input_tokens_seen": 8110304,
      "step": 5030
    },
    {
      "epoch": 0.227936349849476,
      "grad_norm": 0.554357647895813,
      "learning_rate": 4.8071358489796145e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 8117984,
      "step": 5035
    },
    {
      "epoch": 0.22816270173611897,
      "grad_norm": 0.6125796437263489,
      "learning_rate": 4.806757551800665e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 8125472,
      "step": 5040
    },
    {
      "epoch": 0.22838905362276193,
      "grad_norm": 0.7459205985069275,
      "learning_rate": 4.806378898890713e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 8133760,
      "step": 5045
    },
    {
      "epoch": 0.22861540550940493,
      "grad_norm": 0.3204821050167084,
      "learning_rate": 4.80599989030815e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 8141472,
      "step": 5050
    },
    {
      "epoch": 0.2288417573960479,
      "grad_norm": 0.46592041850090027,
      "learning_rate": 4.805620526111426e-05,
      "loss": 0.1992,
      "num_input_tokens_seen": 8149728,
      "step": 5055
    },
    {
      "epoch": 0.22906810928269086,
      "grad_norm": 0.4366299510002136,
      "learning_rate": 4.805240806359042e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 8157408,
      "step": 5060
    },
    {
      "epoch": 0.22929446116933386,
      "grad_norm": 0.7749850749969482,
      "learning_rate": 4.804860731109557e-05,
      "loss": 0.1948,
      "num_input_tokens_seen": 8165920,
      "step": 5065
    },
    {
      "epoch": 0.22952081305597682,
      "grad_norm": 0.3891109824180603,
      "learning_rate": 4.804480300421581e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 8173568,
      "step": 5070
    },
    {
      "epoch": 0.2297471649426198,
      "grad_norm": 0.6970195770263672,
      "learning_rate": 4.804099514353784e-05,
      "loss": 0.2048,
      "num_input_tokens_seen": 8181504,
      "step": 5075
    },
    {
      "epoch": 0.22997351682926276,
      "grad_norm": 0.4040485620498657,
      "learning_rate": 4.8037183729648867e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 8189632,
      "step": 5080
    },
    {
      "epoch": 0.23019986871590575,
      "grad_norm": 0.38815242052078247,
      "learning_rate": 4.803336876313666e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 8198432,
      "step": 5085
    },
    {
      "epoch": 0.23042622060254872,
      "grad_norm": 1.1589335203170776,
      "learning_rate": 4.802955024458953e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 8206208,
      "step": 5090
    },
    {
      "epoch": 0.2306525724891917,
      "grad_norm": 0.9064578413963318,
      "learning_rate": 4.802572817459634e-05,
      "loss": 0.1989,
      "num_input_tokens_seen": 8213952,
      "step": 5095
    },
    {
      "epoch": 0.23087892437583468,
      "grad_norm": 0.3563646674156189,
      "learning_rate": 4.802190255374651e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 8221824,
      "step": 5100
    },
    {
      "epoch": 0.23110527626247765,
      "grad_norm": 0.5691849589347839,
      "learning_rate": 4.801807338263e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 8230464,
      "step": 5105
    },
    {
      "epoch": 0.23133162814912062,
      "grad_norm": 0.3930351734161377,
      "learning_rate": 4.8014240661837306e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 8238784,
      "step": 5110
    },
    {
      "epoch": 0.2315579800357636,
      "grad_norm": 0.6013771891593933,
      "learning_rate": 4.80104043919595e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 8247360,
      "step": 5115
    },
    {
      "epoch": 0.23178433192240658,
      "grad_norm": 0.49210798740386963,
      "learning_rate": 4.800656457358815e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 8255520,
      "step": 5120
    },
    {
      "epoch": 0.23201068380904954,
      "grad_norm": 0.9747103452682495,
      "learning_rate": 4.800272120731544e-05,
      "loss": 0.207,
      "num_input_tokens_seen": 8262848,
      "step": 5125
    },
    {
      "epoch": 0.2322370356956925,
      "grad_norm": 0.4758762717247009,
      "learning_rate": 4.799887429373404e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 8270368,
      "step": 5130
    },
    {
      "epoch": 0.2324633875823355,
      "grad_norm": 0.9492486715316772,
      "learning_rate": 4.79950238334372e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 8278208,
      "step": 5135
    },
    {
      "epoch": 0.23268973946897847,
      "grad_norm": 1.014143943786621,
      "learning_rate": 4.799116982701872e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 8286816,
      "step": 5140
    },
    {
      "epoch": 0.23291609135562144,
      "grad_norm": 0.358458936214447,
      "learning_rate": 4.7987312275072926e-05,
      "loss": 0.2058,
      "num_input_tokens_seen": 8295008,
      "step": 5145
    },
    {
      "epoch": 0.23314244324226444,
      "grad_norm": 1.412386178970337,
      "learning_rate": 4.79834511781947e-05,
      "loss": 0.2214,
      "num_input_tokens_seen": 8302880,
      "step": 5150
    },
    {
      "epoch": 0.2333687951289074,
      "grad_norm": 0.5009074807167053,
      "learning_rate": 4.797958653697947e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 8310880,
      "step": 5155
    },
    {
      "epoch": 0.23359514701555037,
      "grad_norm": 0.5862657427787781,
      "learning_rate": 4.7975718352023225e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 8318944,
      "step": 5160
    },
    {
      "epoch": 0.23382149890219334,
      "grad_norm": 0.6018989682197571,
      "learning_rate": 4.7971846623922476e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 8327040,
      "step": 5165
    },
    {
      "epoch": 0.23404785078883633,
      "grad_norm": 0.5686637759208679,
      "learning_rate": 4.7967971353274294e-05,
      "loss": 0.1815,
      "num_input_tokens_seen": 8334848,
      "step": 5170
    },
    {
      "epoch": 0.2342742026754793,
      "grad_norm": 0.39337170124053955,
      "learning_rate": 4.79640925406763e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 8342464,
      "step": 5175
    },
    {
      "epoch": 0.23450055456212227,
      "grad_norm": 0.5310062766075134,
      "learning_rate": 4.796021018672664e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 8350560,
      "step": 5180
    },
    {
      "epoch": 0.23472690644876526,
      "grad_norm": 0.4395183026790619,
      "learning_rate": 4.795632429202405e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 8358240,
      "step": 5185
    },
    {
      "epoch": 0.23495325833540823,
      "grad_norm": 1.3511202335357666,
      "learning_rate": 4.795243485716775e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 8366464,
      "step": 5190
    },
    {
      "epoch": 0.2351796102220512,
      "grad_norm": 0.4887022376060486,
      "learning_rate": 4.794854188275757e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 8374752,
      "step": 5195
    },
    {
      "epoch": 0.2354059621086942,
      "grad_norm": 0.8251540064811707,
      "learning_rate": 4.794464536939384e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 8382240,
      "step": 5200
    },
    {
      "epoch": 0.2354059621086942,
      "eval_loss": 0.1754050999879837,
      "eval_runtime": 404.8878,
      "eval_samples_per_second": 96.992,
      "eval_steps_per_second": 24.249,
      "num_input_tokens_seen": 8382240,
      "step": 5200
    },
    {
      "epoch": 0.23563231399533716,
      "grad_norm": 0.7284256815910339,
      "learning_rate": 4.794074531767745e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 8390016,
      "step": 5205
    },
    {
      "epoch": 0.23585866588198012,
      "grad_norm": 0.4658472537994385,
      "learning_rate": 4.7936841728209834e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 8398560,
      "step": 5210
    },
    {
      "epoch": 0.2360850177686231,
      "grad_norm": 0.8558294773101807,
      "learning_rate": 4.7932934601593e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 8406528,
      "step": 5215
    },
    {
      "epoch": 0.23631136965526608,
      "grad_norm": 0.8770872950553894,
      "learning_rate": 4.792902393842943e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 8413888,
      "step": 5220
    },
    {
      "epoch": 0.23653772154190905,
      "grad_norm": 0.8042677640914917,
      "learning_rate": 4.792510973932225e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 8421600,
      "step": 5225
    },
    {
      "epoch": 0.23676407342855202,
      "grad_norm": 0.4783654808998108,
      "learning_rate": 4.7921192004875036e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 8429632,
      "step": 5230
    },
    {
      "epoch": 0.236990425315195,
      "grad_norm": 1.1910967826843262,
      "learning_rate": 4.791727073569198e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 8438048,
      "step": 5235
    },
    {
      "epoch": 0.23721677720183798,
      "grad_norm": 0.4067341089248657,
      "learning_rate": 4.7913345932377775e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 8446208,
      "step": 5240
    },
    {
      "epoch": 0.23744312908848095,
      "grad_norm": 0.8732373118400574,
      "learning_rate": 4.790941759553769e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 8454336,
      "step": 5245
    },
    {
      "epoch": 0.23766948097512391,
      "grad_norm": 1.0508780479431152,
      "learning_rate": 4.79054857257775e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 8462848,
      "step": 5250
    },
    {
      "epoch": 0.2378958328617669,
      "grad_norm": 0.6495643258094788,
      "learning_rate": 4.790155032370357e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 8470400,
      "step": 5255
    },
    {
      "epoch": 0.23812218474840988,
      "grad_norm": 0.3850965201854706,
      "learning_rate": 4.789761138992278e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 8478592,
      "step": 5260
    },
    {
      "epoch": 0.23834853663505284,
      "grad_norm": 0.9300469160079956,
      "learning_rate": 4.7893668925042565e-05,
      "loss": 0.1781,
      "num_input_tokens_seen": 8486112,
      "step": 5265
    },
    {
      "epoch": 0.23857488852169584,
      "grad_norm": 0.623296320438385,
      "learning_rate": 4.78897229296709e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 8494112,
      "step": 5270
    },
    {
      "epoch": 0.2388012404083388,
      "grad_norm": 0.5259701609611511,
      "learning_rate": 4.7885773404416315e-05,
      "loss": 0.2306,
      "num_input_tokens_seen": 8502144,
      "step": 5275
    },
    {
      "epoch": 0.23902759229498177,
      "grad_norm": 0.5930491089820862,
      "learning_rate": 4.788182034988786e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 8509952,
      "step": 5280
    },
    {
      "epoch": 0.23925394418162477,
      "grad_norm": 1.4304715394973755,
      "learning_rate": 4.787786376669516e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 8517696,
      "step": 5285
    },
    {
      "epoch": 0.23948029606826773,
      "grad_norm": 0.6945574283599854,
      "learning_rate": 4.787390365544837e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 8525568,
      "step": 5290
    },
    {
      "epoch": 0.2397066479549107,
      "grad_norm": 0.5326865911483765,
      "learning_rate": 4.786994001675818e-05,
      "loss": 0.1951,
      "num_input_tokens_seen": 8534208,
      "step": 5295
    },
    {
      "epoch": 0.23993299984155367,
      "grad_norm": 0.4449228346347809,
      "learning_rate": 4.786597285123584e-05,
      "loss": 0.2157,
      "num_input_tokens_seen": 8542560,
      "step": 5300
    },
    {
      "epoch": 0.24015935172819666,
      "grad_norm": 0.5809661149978638,
      "learning_rate": 4.7862002159493135e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 8550112,
      "step": 5305
    },
    {
      "epoch": 0.24038570361483963,
      "grad_norm": 0.6886299848556519,
      "learning_rate": 4.785802794214239e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 8558496,
      "step": 5310
    },
    {
      "epoch": 0.2406120555014826,
      "grad_norm": 0.5821940302848816,
      "learning_rate": 4.7854050199796495e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 8566816,
      "step": 5315
    },
    {
      "epoch": 0.2408384073881256,
      "grad_norm": 0.38713085651397705,
      "learning_rate": 4.7850068933068845e-05,
      "loss": 0.2129,
      "num_input_tokens_seen": 8574336,
      "step": 5320
    },
    {
      "epoch": 0.24106475927476856,
      "grad_norm": 0.5263111591339111,
      "learning_rate": 4.7846084142573425e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 8581984,
      "step": 5325
    },
    {
      "epoch": 0.24129111116141153,
      "grad_norm": 0.8955719470977783,
      "learning_rate": 4.7842095828924725e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 8589760,
      "step": 5330
    },
    {
      "epoch": 0.2415174630480545,
      "grad_norm": 0.34708020091056824,
      "learning_rate": 4.783810399273779e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 8598496,
      "step": 5335
    },
    {
      "epoch": 0.2417438149346975,
      "grad_norm": 0.4150310754776001,
      "learning_rate": 4.7834108634628226e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 8606048,
      "step": 5340
    },
    {
      "epoch": 0.24197016682134045,
      "grad_norm": 0.6343666315078735,
      "learning_rate": 4.783010975521216e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 8614208,
      "step": 5345
    },
    {
      "epoch": 0.24219651870798342,
      "grad_norm": 0.9839642643928528,
      "learning_rate": 4.782610735510626e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 8622432,
      "step": 5350
    },
    {
      "epoch": 0.24242287059462642,
      "grad_norm": 0.4861735999584198,
      "learning_rate": 4.782210143492776e-05,
      "loss": 0.2165,
      "num_input_tokens_seen": 8630656,
      "step": 5355
    },
    {
      "epoch": 0.24264922248126938,
      "grad_norm": 0.2917002737522125,
      "learning_rate": 4.781809199529442e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 8638592,
      "step": 5360
    },
    {
      "epoch": 0.24287557436791235,
      "grad_norm": 0.4200074374675751,
      "learning_rate": 4.781407903682454e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 8647232,
      "step": 5365
    },
    {
      "epoch": 0.24310192625455534,
      "grad_norm": 0.8186717629432678,
      "learning_rate": 4.781006256013698e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 8655360,
      "step": 5370
    },
    {
      "epoch": 0.2433282781411983,
      "grad_norm": 0.2576800286769867,
      "learning_rate": 4.7806042565851115e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 8663488,
      "step": 5375
    },
    {
      "epoch": 0.24355463002784128,
      "grad_norm": 0.6289105415344238,
      "learning_rate": 4.7802019054586895e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 8671104,
      "step": 5380
    },
    {
      "epoch": 0.24378098191448425,
      "grad_norm": 0.47092923521995544,
      "learning_rate": 4.779799202696479e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 8679104,
      "step": 5385
    },
    {
      "epoch": 0.24400733380112724,
      "grad_norm": 0.9276204109191895,
      "learning_rate": 4.779396148360581e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 8687712,
      "step": 5390
    },
    {
      "epoch": 0.2442336856877702,
      "grad_norm": 0.2983054220676422,
      "learning_rate": 4.7789927425131517e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 8695744,
      "step": 5395
    },
    {
      "epoch": 0.24446003757441317,
      "grad_norm": 0.8342953324317932,
      "learning_rate": 4.778588985216403e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 8703872,
      "step": 5400
    },
    {
      "epoch": 0.24446003757441317,
      "eval_loss": 0.17492441833019257,
      "eval_runtime": 404.8907,
      "eval_samples_per_second": 96.992,
      "eval_steps_per_second": 24.249,
      "num_input_tokens_seen": 8703872,
      "step": 5400
    },
    {
      "epoch": 0.24468638946105617,
      "grad_norm": 0.3011816740036011,
      "learning_rate": 4.778184876532598e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 8711392,
      "step": 5405
    },
    {
      "epoch": 0.24491274134769914,
      "grad_norm": 1.308351993560791,
      "learning_rate": 4.7777804165240556e-05,
      "loss": 0.1844,
      "num_input_tokens_seen": 8719392,
      "step": 5410
    },
    {
      "epoch": 0.2451390932343421,
      "grad_norm": 0.7855434417724609,
      "learning_rate": 4.7773756052531485e-05,
      "loss": 0.198,
      "num_input_tokens_seen": 8727808,
      "step": 5415
    },
    {
      "epoch": 0.24536544512098507,
      "grad_norm": 0.9856816530227661,
      "learning_rate": 4.7769704427823035e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 8737152,
      "step": 5420
    },
    {
      "epoch": 0.24559179700762807,
      "grad_norm": 0.37258315086364746,
      "learning_rate": 4.776564929174003e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 8744960,
      "step": 5425
    },
    {
      "epoch": 0.24581814889427103,
      "grad_norm": 0.36489394307136536,
      "learning_rate": 4.7761590644907806e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 8752960,
      "step": 5430
    },
    {
      "epoch": 0.246044500780914,
      "grad_norm": 0.6333590149879456,
      "learning_rate": 4.7757528487952263e-05,
      "loss": 0.2011,
      "num_input_tokens_seen": 8760800,
      "step": 5435
    },
    {
      "epoch": 0.246270852667557,
      "grad_norm": 1.1110095977783203,
      "learning_rate": 4.7753462821499836e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 8768672,
      "step": 5440
    },
    {
      "epoch": 0.24649720455419996,
      "grad_norm": 0.4032919406890869,
      "learning_rate": 4.774939364617751e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 8776512,
      "step": 5445
    },
    {
      "epoch": 0.24672355644084293,
      "grad_norm": 1.2554502487182617,
      "learning_rate": 4.7745320962612795e-05,
      "loss": 0.2053,
      "num_input_tokens_seen": 8783744,
      "step": 5450
    },
    {
      "epoch": 0.24694990832748592,
      "grad_norm": 0.8689234852790833,
      "learning_rate": 4.7741244771433756e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 8791680,
      "step": 5455
    },
    {
      "epoch": 0.2471762602141289,
      "grad_norm": 0.49534493684768677,
      "learning_rate": 4.7737165073268985e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 8799392,
      "step": 5460
    },
    {
      "epoch": 0.24740261210077186,
      "grad_norm": 0.42027249932289124,
      "learning_rate": 4.7733081868747626e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 8807328,
      "step": 5465
    },
    {
      "epoch": 0.24762896398741482,
      "grad_norm": 0.6971104145050049,
      "learning_rate": 4.772899515849936e-05,
      "loss": 0.2061,
      "num_input_tokens_seen": 8815520,
      "step": 5470
    },
    {
      "epoch": 0.24785531587405782,
      "grad_norm": 0.4381546676158905,
      "learning_rate": 4.7724904943154414e-05,
      "loss": 0.1849,
      "num_input_tokens_seen": 8823104,
      "step": 5475
    },
    {
      "epoch": 0.24808166776070079,
      "grad_norm": 0.816508412361145,
      "learning_rate": 4.772081122334354e-05,
      "loss": 0.1911,
      "num_input_tokens_seen": 8831072,
      "step": 5480
    },
    {
      "epoch": 0.24830801964734375,
      "grad_norm": 0.3724478781223297,
      "learning_rate": 4.771671399969806e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 8838880,
      "step": 5485
    },
    {
      "epoch": 0.24853437153398675,
      "grad_norm": 0.8744020462036133,
      "learning_rate": 4.7712613272849794e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 8847296,
      "step": 5490
    },
    {
      "epoch": 0.24876072342062971,
      "grad_norm": 0.9153328537940979,
      "learning_rate": 4.770850904343114e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 8855008,
      "step": 5495
    },
    {
      "epoch": 0.24898707530727268,
      "grad_norm": 0.49344855546951294,
      "learning_rate": 4.770440131207502e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 8862848,
      "step": 5500
    },
    {
      "epoch": 0.24921342719391565,
      "grad_norm": 0.630511462688446,
      "learning_rate": 4.7700290079414896e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 8870464,
      "step": 5505
    },
    {
      "epoch": 0.24943977908055864,
      "grad_norm": 0.5086699724197388,
      "learning_rate": 4.769617534608477e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 8878080,
      "step": 5510
    },
    {
      "epoch": 0.2496661309672016,
      "grad_norm": 0.5284247994422913,
      "learning_rate": 4.7692057112719193e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 8885984,
      "step": 5515
    },
    {
      "epoch": 0.24989248285384458,
      "grad_norm": 0.48000282049179077,
      "learning_rate": 4.7687935379953234e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 8894016,
      "step": 5520
    },
    {
      "epoch": 0.2501188347404876,
      "grad_norm": 0.6339232325553894,
      "learning_rate": 4.7683810148422534e-05,
      "loss": 0.2058,
      "num_input_tokens_seen": 8902080,
      "step": 5525
    },
    {
      "epoch": 0.2503451866271305,
      "grad_norm": 0.9977306723594666,
      "learning_rate": 4.767968141876324e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 8910336,
      "step": 5530
    },
    {
      "epoch": 0.2505715385137735,
      "grad_norm": 0.6231337189674377,
      "learning_rate": 4.767554919161207e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 8918176,
      "step": 5535
    },
    {
      "epoch": 0.2507978904004165,
      "grad_norm": 0.42167437076568604,
      "learning_rate": 4.767141346760624e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 8926336,
      "step": 5540
    },
    {
      "epoch": 0.25102424228705944,
      "grad_norm": 0.6408863067626953,
      "learning_rate": 4.766727424738356e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 8933728,
      "step": 5545
    },
    {
      "epoch": 0.25125059417370244,
      "grad_norm": 0.6797482371330261,
      "learning_rate": 4.7663131531582325e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 8941472,
      "step": 5550
    },
    {
      "epoch": 0.25147694606034543,
      "grad_norm": 0.6351147890090942,
      "learning_rate": 4.765898532084142e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 8949056,
      "step": 5555
    },
    {
      "epoch": 0.25170329794698837,
      "grad_norm": 0.6289929151535034,
      "learning_rate": 4.765483561580022e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 8956832,
      "step": 5560
    },
    {
      "epoch": 0.25192964983363136,
      "grad_norm": 0.24481436610221863,
      "learning_rate": 4.7650682417098666e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 8965024,
      "step": 5565
    },
    {
      "epoch": 0.25215600172027436,
      "grad_norm": 1.1941330432891846,
      "learning_rate": 4.7646525725377244e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 8973344,
      "step": 5570
    },
    {
      "epoch": 0.2523823536069173,
      "grad_norm": 1.1403477191925049,
      "learning_rate": 4.764236554127696e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 8981344,
      "step": 5575
    },
    {
      "epoch": 0.2526087054935603,
      "grad_norm": 0.783577024936676,
      "learning_rate": 4.7638201865439356e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 8989632,
      "step": 5580
    },
    {
      "epoch": 0.2528350573802033,
      "grad_norm": 0.5155184268951416,
      "learning_rate": 4.7634034698506545e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 8997152,
      "step": 5585
    },
    {
      "epoch": 0.2530614092668462,
      "grad_norm": 0.5093530416488647,
      "learning_rate": 4.762986404112115e-05,
      "loss": 0.1885,
      "num_input_tokens_seen": 9005088,
      "step": 5590
    },
    {
      "epoch": 0.2532877611534892,
      "grad_norm": 1.2222590446472168,
      "learning_rate": 4.762568989392633e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 9013504,
      "step": 5595
    },
    {
      "epoch": 0.2535141130401322,
      "grad_norm": 0.30694255232810974,
      "learning_rate": 4.76215122575658e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 9021760,
      "step": 5600
    },
    {
      "epoch": 0.2535141130401322,
      "eval_loss": 0.17528940737247467,
      "eval_runtime": 403.8425,
      "eval_samples_per_second": 97.243,
      "eval_steps_per_second": 24.311,
      "num_input_tokens_seen": 9021760,
      "step": 5600
    },
    {
      "epoch": 0.25374046492677516,
      "grad_norm": 0.8628082871437073,
      "learning_rate": 4.7617331132683795e-05,
      "loss": 0.2132,
      "num_input_tokens_seen": 9029696,
      "step": 5605
    },
    {
      "epoch": 0.25396681681341815,
      "grad_norm": 0.8646847009658813,
      "learning_rate": 4.7613146519925105e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 9037664,
      "step": 5610
    },
    {
      "epoch": 0.2541931687000611,
      "grad_norm": 0.7009009718894958,
      "learning_rate": 4.7608958419935045e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 9045856,
      "step": 5615
    },
    {
      "epoch": 0.2544195205867041,
      "grad_norm": 0.4357256293296814,
      "learning_rate": 4.760476683335948e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 9054240,
      "step": 5620
    },
    {
      "epoch": 0.2546458724733471,
      "grad_norm": 0.9601346254348755,
      "learning_rate": 4.760057176084479e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 9062752,
      "step": 5625
    },
    {
      "epoch": 0.25487222435999,
      "grad_norm": 0.24299894273281097,
      "learning_rate": 4.759637320303793e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 9071136,
      "step": 5630
    },
    {
      "epoch": 0.255098576246633,
      "grad_norm": 0.6601747870445251,
      "learning_rate": 4.759217116058635e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 9079136,
      "step": 5635
    },
    {
      "epoch": 0.255324928133276,
      "grad_norm": 1.2282837629318237,
      "learning_rate": 4.758796563413807e-05,
      "loss": 0.191,
      "num_input_tokens_seen": 9086784,
      "step": 5640
    },
    {
      "epoch": 0.25555128001991895,
      "grad_norm": 0.5818257927894592,
      "learning_rate": 4.758375662434163e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 9094944,
      "step": 5645
    },
    {
      "epoch": 0.25577763190656194,
      "grad_norm": 0.4419325590133667,
      "learning_rate": 4.7579544131846114e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 9103200,
      "step": 5650
    },
    {
      "epoch": 0.25600398379320494,
      "grad_norm": 0.46842095255851746,
      "learning_rate": 4.757532815730114e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 9110784,
      "step": 5655
    },
    {
      "epoch": 0.2562303356798479,
      "grad_norm": 0.6652507781982422,
      "learning_rate": 4.7571108701356865e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 9118848,
      "step": 5660
    },
    {
      "epoch": 0.25645668756649087,
      "grad_norm": 0.681731104850769,
      "learning_rate": 4.756688576466398e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 9126240,
      "step": 5665
    },
    {
      "epoch": 0.25668303945313387,
      "grad_norm": 0.7158213257789612,
      "learning_rate": 4.756265934787372e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 9134080,
      "step": 5670
    },
    {
      "epoch": 0.2569093913397768,
      "grad_norm": 0.9790230989456177,
      "learning_rate": 4.755842945163785e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 9141792,
      "step": 5675
    },
    {
      "epoch": 0.2571357432264198,
      "grad_norm": 0.7142602801322937,
      "learning_rate": 4.755419607660867e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 9149984,
      "step": 5680
    },
    {
      "epoch": 0.2573620951130628,
      "grad_norm": 0.561445415019989,
      "learning_rate": 4.7549959223439016e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 9157696,
      "step": 5685
    },
    {
      "epoch": 0.25758844699970573,
      "grad_norm": 0.6630258560180664,
      "learning_rate": 4.754571889278228e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 9165600,
      "step": 5690
    },
    {
      "epoch": 0.25781479888634873,
      "grad_norm": 0.7488945722579956,
      "learning_rate": 4.754147508529235e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 9174016,
      "step": 5695
    },
    {
      "epoch": 0.25804115077299167,
      "grad_norm": 0.3451484441757202,
      "learning_rate": 4.75372278016237e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 9182496,
      "step": 5700
    },
    {
      "epoch": 0.25826750265963466,
      "grad_norm": 1.2590445280075073,
      "learning_rate": 4.753297704243129e-05,
      "loss": 0.1898,
      "num_input_tokens_seen": 9190272,
      "step": 5705
    },
    {
      "epoch": 0.25849385454627766,
      "grad_norm": 0.6271164417266846,
      "learning_rate": 4.752872280837066e-05,
      "loss": 0.2084,
      "num_input_tokens_seen": 9198336,
      "step": 5710
    },
    {
      "epoch": 0.2587202064329206,
      "grad_norm": 1.3669573068618774,
      "learning_rate": 4.752446510009786e-05,
      "loss": 0.1794,
      "num_input_tokens_seen": 9206464,
      "step": 5715
    },
    {
      "epoch": 0.2589465583195636,
      "grad_norm": 0.3453759253025055,
      "learning_rate": 4.7520203918269476e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 9214656,
      "step": 5720
    },
    {
      "epoch": 0.2591729102062066,
      "grad_norm": 0.699504017829895,
      "learning_rate": 4.751593926354265e-05,
      "loss": 0.2306,
      "num_input_tokens_seen": 9223072,
      "step": 5725
    },
    {
      "epoch": 0.2593992620928495,
      "grad_norm": 0.36230164766311646,
      "learning_rate": 4.751167113657503e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 9230656,
      "step": 5730
    },
    {
      "epoch": 0.2596256139794925,
      "grad_norm": 0.8676764965057373,
      "learning_rate": 4.7507399538024834e-05,
      "loss": 0.2138,
      "num_input_tokens_seen": 9238944,
      "step": 5735
    },
    {
      "epoch": 0.2598519658661355,
      "grad_norm": 0.34779655933380127,
      "learning_rate": 4.750312446855077e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 9247200,
      "step": 5740
    },
    {
      "epoch": 0.26007831775277845,
      "grad_norm": 0.9929307699203491,
      "learning_rate": 4.749884592881212e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 9254944,
      "step": 5745
    },
    {
      "epoch": 0.26030466963942145,
      "grad_norm": 0.5861169099807739,
      "learning_rate": 4.74945639194687e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 9262496,
      "step": 5750
    },
    {
      "epoch": 0.26053102152606444,
      "grad_norm": 0.680194616317749,
      "learning_rate": 4.749027844118083e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 9270144,
      "step": 5755
    },
    {
      "epoch": 0.2607573734127074,
      "grad_norm": 1.0571250915527344,
      "learning_rate": 4.7485989494609395e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 9278848,
      "step": 5760
    },
    {
      "epoch": 0.2609837252993504,
      "grad_norm": 0.4647243916988373,
      "learning_rate": 4.748169708041581e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 9286976,
      "step": 5765
    },
    {
      "epoch": 0.2612100771859934,
      "grad_norm": 1.023678183555603,
      "learning_rate": 4.7477401199262004e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 9295040,
      "step": 5770
    },
    {
      "epoch": 0.2614364290726363,
      "grad_norm": 1.0223883390426636,
      "learning_rate": 4.747310185181048e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 9302976,
      "step": 5775
    },
    {
      "epoch": 0.2616627809592793,
      "grad_norm": 0.42693018913269043,
      "learning_rate": 4.746879903872422e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 9311488,
      "step": 5780
    },
    {
      "epoch": 0.26188913284592225,
      "grad_norm": 1.0505000352859497,
      "learning_rate": 4.746449276066679e-05,
      "loss": 0.2272,
      "num_input_tokens_seen": 9319008,
      "step": 5785
    },
    {
      "epoch": 0.26211548473256524,
      "grad_norm": 0.8058983683586121,
      "learning_rate": 4.746018301830227e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 9327264,
      "step": 5790
    },
    {
      "epoch": 0.26234183661920824,
      "grad_norm": 0.5145663022994995,
      "learning_rate": 4.7455869812295275e-05,
      "loss": 0.2121,
      "num_input_tokens_seen": 9335424,
      "step": 5795
    },
    {
      "epoch": 0.2625681885058512,
      "grad_norm": 0.7432057857513428,
      "learning_rate": 4.7451553143310964e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 9343744,
      "step": 5800
    },
    {
      "epoch": 0.2625681885058512,
      "eval_loss": 0.17265093326568604,
      "eval_runtime": 404.3175,
      "eval_samples_per_second": 97.129,
      "eval_steps_per_second": 24.283,
      "num_input_tokens_seen": 9343744,
      "step": 5800
    },
    {
      "epoch": 0.26279454039249417,
      "grad_norm": 0.40748724341392517,
      "learning_rate": 4.744723301201501e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 9351456,
      "step": 5805
    },
    {
      "epoch": 0.26302089227913716,
      "grad_norm": 0.9156345129013062,
      "learning_rate": 4.744290941907364e-05,
      "loss": 0.1985,
      "num_input_tokens_seen": 9358944,
      "step": 5810
    },
    {
      "epoch": 0.2632472441657801,
      "grad_norm": 0.8601633310317993,
      "learning_rate": 4.7438582365153594e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 9367008,
      "step": 5815
    },
    {
      "epoch": 0.2634735960524231,
      "grad_norm": 0.693898618221283,
      "learning_rate": 4.743425185092217e-05,
      "loss": 0.2148,
      "num_input_tokens_seen": 9375008,
      "step": 5820
    },
    {
      "epoch": 0.2636999479390661,
      "grad_norm": 0.5600563287734985,
      "learning_rate": 4.742991787704719e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 9382976,
      "step": 5825
    },
    {
      "epoch": 0.26392629982570903,
      "grad_norm": 0.5999929308891296,
      "learning_rate": 4.7425580444196994e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 9391008,
      "step": 5830
    },
    {
      "epoch": 0.264152651712352,
      "grad_norm": 0.545287013053894,
      "learning_rate": 4.742123955304048e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 9399104,
      "step": 5835
    },
    {
      "epoch": 0.264379003598995,
      "grad_norm": 0.4050139784812927,
      "learning_rate": 4.741689520424706e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 9407232,
      "step": 5840
    },
    {
      "epoch": 0.26460535548563796,
      "grad_norm": 0.6832709312438965,
      "learning_rate": 4.741254739848669e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 9415680,
      "step": 5845
    },
    {
      "epoch": 0.26483170737228096,
      "grad_norm": 0.6217070817947388,
      "learning_rate": 4.740819613642987e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 9423936,
      "step": 5850
    },
    {
      "epoch": 0.26505805925892395,
      "grad_norm": 0.44570454955101013,
      "learning_rate": 4.74038414187476e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 9432384,
      "step": 5855
    },
    {
      "epoch": 0.2652844111455669,
      "grad_norm": 0.9275084733963013,
      "learning_rate": 4.739948324611144e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 9440512,
      "step": 5860
    },
    {
      "epoch": 0.2655107630322099,
      "grad_norm": 0.9646974205970764,
      "learning_rate": 4.7395121619193465e-05,
      "loss": 0.1999,
      "num_input_tokens_seen": 9448256,
      "step": 5865
    },
    {
      "epoch": 0.2657371149188528,
      "grad_norm": 0.6389604806900024,
      "learning_rate": 4.7390756538666313e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 9456096,
      "step": 5870
    },
    {
      "epoch": 0.2659634668054958,
      "grad_norm": 1.0727211236953735,
      "learning_rate": 4.738638800520311e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 9463328,
      "step": 5875
    },
    {
      "epoch": 0.2661898186921388,
      "grad_norm": 0.710365355014801,
      "learning_rate": 4.738201601947757e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 9471104,
      "step": 5880
    },
    {
      "epoch": 0.26641617057878175,
      "grad_norm": 0.826835572719574,
      "learning_rate": 4.7377640582163876e-05,
      "loss": 0.2112,
      "num_input_tokens_seen": 9478752,
      "step": 5885
    },
    {
      "epoch": 0.26664252246542475,
      "grad_norm": 0.41567641496658325,
      "learning_rate": 4.7373261693936786e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 9487232,
      "step": 5890
    },
    {
      "epoch": 0.26686887435206774,
      "grad_norm": 0.9625677466392517,
      "learning_rate": 4.7368879355471595e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 9494944,
      "step": 5895
    },
    {
      "epoch": 0.2670952262387107,
      "grad_norm": 1.0546866655349731,
      "learning_rate": 4.736449356744409e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 9502976,
      "step": 5900
    },
    {
      "epoch": 0.2673215781253537,
      "grad_norm": 0.46502307057380676,
      "learning_rate": 4.736010433053064e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 9511296,
      "step": 5905
    },
    {
      "epoch": 0.26754793001199667,
      "grad_norm": 0.32968175411224365,
      "learning_rate": 4.73557116454081e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 9519168,
      "step": 5910
    },
    {
      "epoch": 0.2677742818986396,
      "grad_norm": 0.6044268608093262,
      "learning_rate": 4.735131551275389e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 9527328,
      "step": 5915
    },
    {
      "epoch": 0.2680006337852826,
      "grad_norm": 0.6576837301254272,
      "learning_rate": 4.734691593324594e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 9535104,
      "step": 5920
    },
    {
      "epoch": 0.2682269856719256,
      "grad_norm": 0.5958639979362488,
      "learning_rate": 4.734251290756272e-05,
      "loss": 0.218,
      "num_input_tokens_seen": 9542368,
      "step": 5925
    },
    {
      "epoch": 0.26845333755856854,
      "grad_norm": 0.37975209951400757,
      "learning_rate": 4.7338106436383246e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 9550080,
      "step": 5930
    },
    {
      "epoch": 0.26867968944521153,
      "grad_norm": 0.6686599254608154,
      "learning_rate": 4.733369652038703e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 9558720,
      "step": 5935
    },
    {
      "epoch": 0.26890604133185453,
      "grad_norm": 0.4212051331996918,
      "learning_rate": 4.7329283160254156e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 9566464,
      "step": 5940
    },
    {
      "epoch": 0.26913239321849747,
      "grad_norm": 0.6723652482032776,
      "learning_rate": 4.732486635666521e-05,
      "loss": 0.2047,
      "num_input_tokens_seen": 9574304,
      "step": 5945
    },
    {
      "epoch": 0.26935874510514046,
      "grad_norm": 1.1781935691833496,
      "learning_rate": 4.732044611030132e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 9582016,
      "step": 5950
    },
    {
      "epoch": 0.2695850969917834,
      "grad_norm": 0.35691994428634644,
      "learning_rate": 4.731602242184414e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 9590112,
      "step": 5955
    },
    {
      "epoch": 0.2698114488784264,
      "grad_norm": 0.9523367881774902,
      "learning_rate": 4.7311595291975864e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 9597920,
      "step": 5960
    },
    {
      "epoch": 0.2700378007650694,
      "grad_norm": 0.47833529114723206,
      "learning_rate": 4.7307164721379216e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 9605824,
      "step": 5965
    },
    {
      "epoch": 0.27026415265171233,
      "grad_norm": 0.8762959837913513,
      "learning_rate": 4.730273071073743e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 9614208,
      "step": 5970
    },
    {
      "epoch": 0.2704905045383553,
      "grad_norm": 0.6768898367881775,
      "learning_rate": 4.729829326073429e-05,
      "loss": 0.1853,
      "num_input_tokens_seen": 9622208,
      "step": 5975
    },
    {
      "epoch": 0.2707168564249983,
      "grad_norm": 0.7478813529014587,
      "learning_rate": 4.7293852372054126e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 9630208,
      "step": 5980
    },
    {
      "epoch": 0.27094320831164126,
      "grad_norm": 1.0893343687057495,
      "learning_rate": 4.728940804538176e-05,
      "loss": 0.2097,
      "num_input_tokens_seen": 9638240,
      "step": 5985
    },
    {
      "epoch": 0.27116956019828425,
      "grad_norm": 0.7069440484046936,
      "learning_rate": 4.7284960281402556e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 9646560,
      "step": 5990
    },
    {
      "epoch": 0.27139591208492725,
      "grad_norm": 0.7813951373100281,
      "learning_rate": 4.728050908080244e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 9655136,
      "step": 5995
    },
    {
      "epoch": 0.2716222639715702,
      "grad_norm": 0.9443720579147339,
      "learning_rate": 4.727605444426782e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 9663168,
      "step": 6000
    },
    {
      "epoch": 0.2716222639715702,
      "eval_loss": 0.17137756943702698,
      "eval_runtime": 403.8958,
      "eval_samples_per_second": 97.231,
      "eval_steps_per_second": 24.308,
      "num_input_tokens_seen": 9663168,
      "step": 6000
    },
    {
      "epoch": 0.2718486158582132,
      "grad_norm": 0.4658046364784241,
      "learning_rate": 4.727159637248567e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 9671008,
      "step": 6005
    },
    {
      "epoch": 0.2720749677448562,
      "grad_norm": 0.5398953557014465,
      "learning_rate": 4.7267134866143474e-05,
      "loss": 0.1815,
      "num_input_tokens_seen": 9679136,
      "step": 6010
    },
    {
      "epoch": 0.2723013196314991,
      "grad_norm": 0.3514156937599182,
      "learning_rate": 4.726266992592926e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 9687392,
      "step": 6015
    },
    {
      "epoch": 0.2725276715181421,
      "grad_norm": 0.6699734330177307,
      "learning_rate": 4.725820155253157e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 9695424,
      "step": 6020
    },
    {
      "epoch": 0.2727540234047851,
      "grad_norm": 1.1323822736740112,
      "learning_rate": 4.725372974663948e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 9703360,
      "step": 6025
    },
    {
      "epoch": 0.27298037529142805,
      "grad_norm": 0.9669657349586487,
      "learning_rate": 4.724925450894262e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 9710752,
      "step": 6030
    },
    {
      "epoch": 0.27320672717807104,
      "grad_norm": 0.678256094455719,
      "learning_rate": 4.72447758401311e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 9718848,
      "step": 6035
    },
    {
      "epoch": 0.273433079064714,
      "grad_norm": 0.6179534196853638,
      "learning_rate": 4.7240293740895616e-05,
      "loss": 0.1868,
      "num_input_tokens_seen": 9726752,
      "step": 6040
    },
    {
      "epoch": 0.273659430951357,
      "grad_norm": 0.9689708948135376,
      "learning_rate": 4.723580821192733e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 9735520,
      "step": 6045
    },
    {
      "epoch": 0.27388578283799997,
      "grad_norm": 0.45306912064552307,
      "learning_rate": 4.7231319253917996e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 9743456,
      "step": 6050
    },
    {
      "epoch": 0.2741121347246429,
      "grad_norm": 0.8087562918663025,
      "learning_rate": 4.722682686755986e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 9751744,
      "step": 6055
    },
    {
      "epoch": 0.2743384866112859,
      "grad_norm": 0.6167035698890686,
      "learning_rate": 4.722233105354569e-05,
      "loss": 0.213,
      "num_input_tokens_seen": 9760128,
      "step": 6060
    },
    {
      "epoch": 0.2745648384979289,
      "grad_norm": 0.9357110261917114,
      "learning_rate": 4.7217831812568815e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 9768736,
      "step": 6065
    },
    {
      "epoch": 0.27479119038457184,
      "grad_norm": 0.2802063822746277,
      "learning_rate": 4.721332914532307e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 9776864,
      "step": 6070
    },
    {
      "epoch": 0.27501754227121483,
      "grad_norm": 0.6155784130096436,
      "learning_rate": 4.720882305250281e-05,
      "loss": 0.1885,
      "num_input_tokens_seen": 9785056,
      "step": 6075
    },
    {
      "epoch": 0.2752438941578578,
      "grad_norm": 0.49853792786598206,
      "learning_rate": 4.720431353480295e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 9792928,
      "step": 6080
    },
    {
      "epoch": 0.27547024604450077,
      "grad_norm": 0.5441297888755798,
      "learning_rate": 4.719980059291891e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 9801248,
      "step": 6085
    },
    {
      "epoch": 0.27569659793114376,
      "grad_norm": 0.6498478651046753,
      "learning_rate": 4.7195284227546634e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 9809280,
      "step": 6090
    },
    {
      "epoch": 0.27592294981778676,
      "grad_norm": 0.5733804106712341,
      "learning_rate": 4.7190764439382604e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 9816672,
      "step": 6095
    },
    {
      "epoch": 0.2761493017044297,
      "grad_norm": 0.31964734196662903,
      "learning_rate": 4.7186241229123826e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 9824192,
      "step": 6100
    },
    {
      "epoch": 0.2763756535910727,
      "grad_norm": 0.6099514365196228,
      "learning_rate": 4.718171459746785e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 9832384,
      "step": 6105
    },
    {
      "epoch": 0.2766020054777157,
      "grad_norm": 0.8476884961128235,
      "learning_rate": 4.717718454511273e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 9839872,
      "step": 6110
    },
    {
      "epoch": 0.2768283573643586,
      "grad_norm": 0.7319749593734741,
      "learning_rate": 4.7172651072757056e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 9847808,
      "step": 6115
    },
    {
      "epoch": 0.2770547092510016,
      "grad_norm": 0.40302255749702454,
      "learning_rate": 4.7168114181099945e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 9856256,
      "step": 6120
    },
    {
      "epoch": 0.27728106113764456,
      "grad_norm": 0.6141369342803955,
      "learning_rate": 4.716357387084105e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 9863456,
      "step": 6125
    },
    {
      "epoch": 0.27750741302428755,
      "grad_norm": 0.45158249139785767,
      "learning_rate": 4.715903014268054e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 9870912,
      "step": 6130
    },
    {
      "epoch": 0.27773376491093055,
      "grad_norm": 0.44462865591049194,
      "learning_rate": 4.715448299731911e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 9878336,
      "step": 6135
    },
    {
      "epoch": 0.2779601167975735,
      "grad_norm": 0.9948895573616028,
      "learning_rate": 4.7149932435457986e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 9887392,
      "step": 6140
    },
    {
      "epoch": 0.2781864686842165,
      "grad_norm": 0.7475268840789795,
      "learning_rate": 4.714537845779894e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 9895520,
      "step": 6145
    },
    {
      "epoch": 0.2784128205708595,
      "grad_norm": 0.470700740814209,
      "learning_rate": 4.714082106504423e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 9903104,
      "step": 6150
    },
    {
      "epoch": 0.2786391724575024,
      "grad_norm": 0.748724102973938,
      "learning_rate": 4.713626025789667e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 9910976,
      "step": 6155
    },
    {
      "epoch": 0.2788655243441454,
      "grad_norm": 0.5394397974014282,
      "learning_rate": 4.7131696037059606e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 9918944,
      "step": 6160
    },
    {
      "epoch": 0.2790918762307884,
      "grad_norm": 1.1438031196594238,
      "learning_rate": 4.712712840323689e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 9927328,
      "step": 6165
    },
    {
      "epoch": 0.27931822811743134,
      "grad_norm": 0.7849851250648499,
      "learning_rate": 4.71225573571329e-05,
      "loss": 0.1973,
      "num_input_tokens_seen": 9935200,
      "step": 6170
    },
    {
      "epoch": 0.27954458000407434,
      "grad_norm": 0.5369264483451843,
      "learning_rate": 4.711798289945256e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 9943136,
      "step": 6175
    },
    {
      "epoch": 0.27977093189071733,
      "grad_norm": 0.7446088790893555,
      "learning_rate": 4.71134050309013e-05,
      "loss": 0.1837,
      "num_input_tokens_seen": 9951584,
      "step": 6180
    },
    {
      "epoch": 0.2799972837773603,
      "grad_norm": 0.4031681418418884,
      "learning_rate": 4.710882375218509e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 9959232,
      "step": 6185
    },
    {
      "epoch": 0.28022363566400327,
      "grad_norm": 0.4120524525642395,
      "learning_rate": 4.7104239064010424e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 9967424,
      "step": 6190
    },
    {
      "epoch": 0.2804499875506462,
      "grad_norm": 0.4257764518260956,
      "learning_rate": 4.709965096708432e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 9975328,
      "step": 6195
    },
    {
      "epoch": 0.2806763394372892,
      "grad_norm": 0.7966869473457336,
      "learning_rate": 4.709505946211431e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 9983680,
      "step": 6200
    },
    {
      "epoch": 0.2806763394372892,
      "eval_loss": 0.1708231419324875,
      "eval_runtime": 404.5193,
      "eval_samples_per_second": 97.081,
      "eval_steps_per_second": 24.271,
      "num_input_tokens_seen": 9983680,
      "step": 6200
    },
    {
      "epoch": 0.2809026913239322,
      "grad_norm": 0.6780665516853333,
      "learning_rate": 4.709046454980846e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 9991104,
      "step": 6205
    },
    {
      "epoch": 0.28112904321057514,
      "grad_norm": 0.6198112368583679,
      "learning_rate": 4.708586623087538e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 9998848,
      "step": 6210
    },
    {
      "epoch": 0.28135539509721813,
      "grad_norm": 0.40498557686805725,
      "learning_rate": 4.708126450602418e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 10007072,
      "step": 6215
    },
    {
      "epoch": 0.2815817469838611,
      "grad_norm": 0.5124131441116333,
      "learning_rate": 4.7076659375964495e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 10014976,
      "step": 6220
    },
    {
      "epoch": 0.28180809887050406,
      "grad_norm": 0.8006425499916077,
      "learning_rate": 4.707205084140651e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 10022656,
      "step": 6225
    },
    {
      "epoch": 0.28203445075714706,
      "grad_norm": 0.5800185799598694,
      "learning_rate": 4.7067438903060904e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 10031072,
      "step": 6230
    },
    {
      "epoch": 0.28226080264379005,
      "grad_norm": 0.8686108589172363,
      "learning_rate": 4.70628235616389e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 10038912,
      "step": 6235
    },
    {
      "epoch": 0.282487154530433,
      "grad_norm": 0.6274464726448059,
      "learning_rate": 4.7058204817852256e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 10047008,
      "step": 6240
    },
    {
      "epoch": 0.282713506417076,
      "grad_norm": 0.4381249248981476,
      "learning_rate": 4.705358267241322e-05,
      "loss": 0.1852,
      "num_input_tokens_seen": 10054880,
      "step": 6245
    },
    {
      "epoch": 0.282939858303719,
      "grad_norm": 0.6135722994804382,
      "learning_rate": 4.704895712603459e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 10062656,
      "step": 6250
    },
    {
      "epoch": 0.2831662101903619,
      "grad_norm": 1.2599575519561768,
      "learning_rate": 4.704432817942969e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 10071104,
      "step": 6255
    },
    {
      "epoch": 0.2833925620770049,
      "grad_norm": 1.6461946964263916,
      "learning_rate": 4.703969583331236e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 10079072,
      "step": 6260
    },
    {
      "epoch": 0.2836189139636479,
      "grad_norm": 0.9381948113441467,
      "learning_rate": 4.7035060088396965e-05,
      "loss": 0.1998,
      "num_input_tokens_seen": 10087456,
      "step": 6265
    },
    {
      "epoch": 0.28384526585029085,
      "grad_norm": 0.95367830991745,
      "learning_rate": 4.703042094539839e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 10095424,
      "step": 6270
    },
    {
      "epoch": 0.28407161773693385,
      "grad_norm": 1.0527158975601196,
      "learning_rate": 4.702577840503206e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 10103168,
      "step": 6275
    },
    {
      "epoch": 0.2842979696235768,
      "grad_norm": 0.9658971428871155,
      "learning_rate": 4.70211324680139e-05,
      "loss": 0.1214,
      "num_input_tokens_seen": 10111168,
      "step": 6280
    },
    {
      "epoch": 0.2845243215102198,
      "grad_norm": 0.4593041241168976,
      "learning_rate": 4.7016483135060386e-05,
      "loss": 0.1954,
      "num_input_tokens_seen": 10119296,
      "step": 6285
    },
    {
      "epoch": 0.2847506733968628,
      "grad_norm": 0.4959751069545746,
      "learning_rate": 4.701183040688849e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 10127744,
      "step": 6290
    },
    {
      "epoch": 0.2849770252835057,
      "grad_norm": 0.6325576901435852,
      "learning_rate": 4.700717428421573e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 10135872,
      "step": 6295
    },
    {
      "epoch": 0.2852033771701487,
      "grad_norm": 0.5069641470909119,
      "learning_rate": 4.700251476776014e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 10144256,
      "step": 6300
    },
    {
      "epoch": 0.2854297290567917,
      "grad_norm": 0.5335239171981812,
      "learning_rate": 4.699785185824026e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 10152672,
      "step": 6305
    },
    {
      "epoch": 0.28565608094343464,
      "grad_norm": 0.6658104062080383,
      "learning_rate": 4.699318555637519e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 10160608,
      "step": 6310
    },
    {
      "epoch": 0.28588243283007764,
      "grad_norm": 1.197942852973938,
      "learning_rate": 4.6988515862884525e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 10168608,
      "step": 6315
    },
    {
      "epoch": 0.28610878471672063,
      "grad_norm": 0.5101622343063354,
      "learning_rate": 4.698384277848838e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 10176320,
      "step": 6320
    },
    {
      "epoch": 0.28633513660336357,
      "grad_norm": 0.43001577258110046,
      "learning_rate": 4.6979166303907425e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 10183840,
      "step": 6325
    },
    {
      "epoch": 0.28656148849000657,
      "grad_norm": 0.5841711759567261,
      "learning_rate": 4.697448643986281e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 10191936,
      "step": 6330
    },
    {
      "epoch": 0.28678784037664956,
      "grad_norm": 0.7264339327812195,
      "learning_rate": 4.696980318707624e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 10199392,
      "step": 6335
    },
    {
      "epoch": 0.2870141922632925,
      "grad_norm": 0.5021238923072815,
      "learning_rate": 4.6965116546269924e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 10207200,
      "step": 6340
    },
    {
      "epoch": 0.2872405441499355,
      "grad_norm": 0.5270203948020935,
      "learning_rate": 4.6960426518166615e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 10214880,
      "step": 6345
    },
    {
      "epoch": 0.2874668960365785,
      "grad_norm": 0.5381638407707214,
      "learning_rate": 4.6955733103489556e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 10223968,
      "step": 6350
    },
    {
      "epoch": 0.28769324792322143,
      "grad_norm": 0.8376690745353699,
      "learning_rate": 4.695103630296255e-05,
      "loss": 0.1844,
      "num_input_tokens_seen": 10231776,
      "step": 6355
    },
    {
      "epoch": 0.2879195998098644,
      "grad_norm": 0.806469738483429,
      "learning_rate": 4.694633611730988e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 10240160,
      "step": 6360
    },
    {
      "epoch": 0.28814595169650736,
      "grad_norm": 1.195651650428772,
      "learning_rate": 4.694163254725639e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 10248288,
      "step": 6365
    },
    {
      "epoch": 0.28837230358315036,
      "grad_norm": 0.4341840445995331,
      "learning_rate": 4.693692559352743e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 10255968,
      "step": 6370
    },
    {
      "epoch": 0.28859865546979335,
      "grad_norm": 0.44151580333709717,
      "learning_rate": 4.693221525684886e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 10263712,
      "step": 6375
    },
    {
      "epoch": 0.2888250073564363,
      "grad_norm": 0.4381527900695801,
      "learning_rate": 4.6927501537947084e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 10271488,
      "step": 6380
    },
    {
      "epoch": 0.2890513592430793,
      "grad_norm": 1.0793476104736328,
      "learning_rate": 4.692278443754901e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 10279968,
      "step": 6385
    },
    {
      "epoch": 0.2892777111297223,
      "grad_norm": 1.1136184930801392,
      "learning_rate": 4.691806395638208e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 10288672,
      "step": 6390
    },
    {
      "epoch": 0.2895040630163652,
      "grad_norm": 0.5135229825973511,
      "learning_rate": 4.6913340095174255e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 10297344,
      "step": 6395
    },
    {
      "epoch": 0.2897304149030082,
      "grad_norm": 0.5260220170021057,
      "learning_rate": 4.690861285465399e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 10305504,
      "step": 6400
    },
    {
      "epoch": 0.2897304149030082,
      "eval_loss": 0.17028893530368805,
      "eval_runtime": 404.2954,
      "eval_samples_per_second": 97.134,
      "eval_steps_per_second": 24.284,
      "num_input_tokens_seen": 10305504,
      "step": 6400
    },
    {
      "epoch": 0.2899567667896512,
      "grad_norm": 0.6331255435943604,
      "learning_rate": 4.690388223555031e-05,
      "loss": 0.2198,
      "num_input_tokens_seen": 10314208,
      "step": 6405
    },
    {
      "epoch": 0.29018311867629415,
      "grad_norm": 0.8553824424743652,
      "learning_rate": 4.689914823859273e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 10322944,
      "step": 6410
    },
    {
      "epoch": 0.29040947056293714,
      "grad_norm": 0.9351500272750854,
      "learning_rate": 4.689441086451129e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 10331488,
      "step": 6415
    },
    {
      "epoch": 0.29063582244958014,
      "grad_norm": 0.7550642490386963,
      "learning_rate": 4.688967011403655e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 10339904,
      "step": 6420
    },
    {
      "epoch": 0.2908621743362231,
      "grad_norm": 0.5370323061943054,
      "learning_rate": 4.68849259878996e-05,
      "loss": 0.1817,
      "num_input_tokens_seen": 10347840,
      "step": 6425
    },
    {
      "epoch": 0.2910885262228661,
      "grad_norm": 0.7438840270042419,
      "learning_rate": 4.6880178486832036e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 10355552,
      "step": 6430
    },
    {
      "epoch": 0.29131487810950907,
      "grad_norm": 0.958197832107544,
      "learning_rate": 4.687542761156598e-05,
      "loss": 0.1976,
      "num_input_tokens_seen": 10363072,
      "step": 6435
    },
    {
      "epoch": 0.291541229996152,
      "grad_norm": 0.8126586079597473,
      "learning_rate": 4.6870673362834096e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 10370688,
      "step": 6440
    },
    {
      "epoch": 0.291767581882795,
      "grad_norm": 0.364963173866272,
      "learning_rate": 4.6865915741369526e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 10378304,
      "step": 6445
    },
    {
      "epoch": 0.29199393376943794,
      "grad_norm": 0.5753499269485474,
      "learning_rate": 4.686115474790597e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 10386464,
      "step": 6450
    },
    {
      "epoch": 0.29222028565608094,
      "grad_norm": 1.0107486248016357,
      "learning_rate": 4.685639038317762e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 10394144,
      "step": 6455
    },
    {
      "epoch": 0.29244663754272393,
      "grad_norm": 0.7798843383789062,
      "learning_rate": 4.685162264791921e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 10402720,
      "step": 6460
    },
    {
      "epoch": 0.29267298942936687,
      "grad_norm": 0.8983163237571716,
      "learning_rate": 4.684685154286599e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 10410784,
      "step": 6465
    },
    {
      "epoch": 0.29289934131600986,
      "grad_norm": 0.555179238319397,
      "learning_rate": 4.684207706875371e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 10418304,
      "step": 6470
    },
    {
      "epoch": 0.29312569320265286,
      "grad_norm": 0.7475776672363281,
      "learning_rate": 4.683729922631866e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 10426272,
      "step": 6475
    },
    {
      "epoch": 0.2933520450892958,
      "grad_norm": 0.6098474264144897,
      "learning_rate": 4.683251801629765e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 10434112,
      "step": 6480
    },
    {
      "epoch": 0.2935783969759388,
      "grad_norm": 0.4485909044742584,
      "learning_rate": 4.6827733439428e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 10441728,
      "step": 6485
    },
    {
      "epoch": 0.2938047488625818,
      "grad_norm": 0.4764844477176666,
      "learning_rate": 4.682294549644754e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 10449792,
      "step": 6490
    },
    {
      "epoch": 0.2940311007492247,
      "grad_norm": 0.8237794637680054,
      "learning_rate": 4.681815418809464e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 10458208,
      "step": 6495
    },
    {
      "epoch": 0.2942574526358677,
      "grad_norm": 0.6525758504867554,
      "learning_rate": 4.681335951510819e-05,
      "loss": 0.21,
      "num_input_tokens_seen": 10466144,
      "step": 6500
    },
    {
      "epoch": 0.2944838045225107,
      "grad_norm": 0.4244976043701172,
      "learning_rate": 4.6808561478227576e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 10474688,
      "step": 6505
    },
    {
      "epoch": 0.29471015640915366,
      "grad_norm": 0.5868838429450989,
      "learning_rate": 4.680376007819271e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 10482752,
      "step": 6510
    },
    {
      "epoch": 0.29493650829579665,
      "grad_norm": 0.7160568237304688,
      "learning_rate": 4.679895531574405e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 10490976,
      "step": 6515
    },
    {
      "epoch": 0.29516286018243965,
      "grad_norm": 0.4093823730945587,
      "learning_rate": 4.679414719162253e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 10498464,
      "step": 6520
    },
    {
      "epoch": 0.2953892120690826,
      "grad_norm": 0.3921254277229309,
      "learning_rate": 4.6789335706569635e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 10506560,
      "step": 6525
    },
    {
      "epoch": 0.2956155639557256,
      "grad_norm": 0.7541391253471375,
      "learning_rate": 4.678452086132734e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 10514208,
      "step": 6530
    },
    {
      "epoch": 0.2958419158423685,
      "grad_norm": 0.7789086699485779,
      "learning_rate": 4.677970265663818e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 10522560,
      "step": 6535
    },
    {
      "epoch": 0.2960682677290115,
      "grad_norm": 0.3331380784511566,
      "learning_rate": 4.677488109324517e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 10530368,
      "step": 6540
    },
    {
      "epoch": 0.2962946196156545,
      "grad_norm": 0.35327357053756714,
      "learning_rate": 4.6770056171891846e-05,
      "loss": 0.1834,
      "num_input_tokens_seen": 10537920,
      "step": 6545
    },
    {
      "epoch": 0.29652097150229745,
      "grad_norm": 0.3553358316421509,
      "learning_rate": 4.6765227893322286e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 10545824,
      "step": 6550
    },
    {
      "epoch": 0.29674732338894044,
      "grad_norm": 1.8299978971481323,
      "learning_rate": 4.676039625828107e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 10554016,
      "step": 6555
    },
    {
      "epoch": 0.29697367527558344,
      "grad_norm": 0.4967997074127197,
      "learning_rate": 4.675556126751328e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 10562464,
      "step": 6560
    },
    {
      "epoch": 0.2972000271622264,
      "grad_norm": 0.5803517699241638,
      "learning_rate": 4.6750722921764556e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 10571008,
      "step": 6565
    },
    {
      "epoch": 0.29742637904886937,
      "grad_norm": 1.2557320594787598,
      "learning_rate": 4.674588122178102e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 10579520,
      "step": 6570
    },
    {
      "epoch": 0.29765273093551237,
      "grad_norm": 0.368050217628479,
      "learning_rate": 4.674103616830931e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 10588192,
      "step": 6575
    },
    {
      "epoch": 0.2978790828221553,
      "grad_norm": 0.7617861032485962,
      "learning_rate": 4.673618776209663e-05,
      "loss": 0.2067,
      "num_input_tokens_seen": 10596512,
      "step": 6580
    },
    {
      "epoch": 0.2981054347087983,
      "grad_norm": 0.5569435954093933,
      "learning_rate": 4.673133600389063e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 10603872,
      "step": 6585
    },
    {
      "epoch": 0.2983317865954413,
      "grad_norm": 0.6230464577674866,
      "learning_rate": 4.672648089443953e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 10611808,
      "step": 6590
    },
    {
      "epoch": 0.29855813848208423,
      "grad_norm": 0.6157615184783936,
      "learning_rate": 4.672162243449204e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 10620416,
      "step": 6595
    },
    {
      "epoch": 0.29878449036872723,
      "grad_norm": 0.6098151206970215,
      "learning_rate": 4.67167606247974e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 10628064,
      "step": 6600
    },
    {
      "epoch": 0.29878449036872723,
      "eval_loss": 0.1687159687280655,
      "eval_runtime": 404.0392,
      "eval_samples_per_second": 97.196,
      "eval_steps_per_second": 24.3,
      "num_input_tokens_seen": 10628064,
      "step": 6600
    },
    {
      "epoch": 0.2990108422553702,
      "grad_norm": 0.25682970881462097,
      "learning_rate": 4.671189546610536e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 10635968,
      "step": 6605
    },
    {
      "epoch": 0.29923719414201316,
      "grad_norm": 0.8740496635437012,
      "learning_rate": 4.67070269591662e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 10644224,
      "step": 6610
    },
    {
      "epoch": 0.29946354602865616,
      "grad_norm": 1.1657347679138184,
      "learning_rate": 4.670215510473068e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 10652416,
      "step": 6615
    },
    {
      "epoch": 0.2996898979152991,
      "grad_norm": 0.8153241872787476,
      "learning_rate": 4.669727990355013e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 10659904,
      "step": 6620
    },
    {
      "epoch": 0.2999162498019421,
      "grad_norm": 0.4676041603088379,
      "learning_rate": 4.669240135637635e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 10668096,
      "step": 6625
    },
    {
      "epoch": 0.3001426016885851,
      "grad_norm": 0.636280357837677,
      "learning_rate": 4.6687519463961675e-05,
      "loss": 0.2273,
      "num_input_tokens_seen": 10676000,
      "step": 6630
    },
    {
      "epoch": 0.300368953575228,
      "grad_norm": 0.5961371064186096,
      "learning_rate": 4.668263422705896e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 10683744,
      "step": 6635
    },
    {
      "epoch": 0.300595305461871,
      "grad_norm": 0.5240316390991211,
      "learning_rate": 4.667774564642156e-05,
      "loss": 0.1862,
      "num_input_tokens_seen": 10691456,
      "step": 6640
    },
    {
      "epoch": 0.300821657348514,
      "grad_norm": 0.2767482101917267,
      "learning_rate": 4.6672853722803365e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 10699200,
      "step": 6645
    },
    {
      "epoch": 0.30104800923515695,
      "grad_norm": 0.7537439465522766,
      "learning_rate": 4.666795845695877e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 10706752,
      "step": 6650
    },
    {
      "epoch": 0.30127436112179995,
      "grad_norm": 0.346945583820343,
      "learning_rate": 4.666305984964269e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 10715360,
      "step": 6655
    },
    {
      "epoch": 0.30150071300844294,
      "grad_norm": 0.6247114539146423,
      "learning_rate": 4.6658157901610535e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 10723872,
      "step": 6660
    },
    {
      "epoch": 0.3017270648950859,
      "grad_norm": 0.5466428995132446,
      "learning_rate": 4.665325261361826e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 10732128,
      "step": 6665
    },
    {
      "epoch": 0.3019534167817289,
      "grad_norm": 0.7220366597175598,
      "learning_rate": 4.664834398642232e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 10740480,
      "step": 6670
    },
    {
      "epoch": 0.3021797686683719,
      "grad_norm": 0.7670433521270752,
      "learning_rate": 4.6643432020779686e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 10748128,
      "step": 6675
    },
    {
      "epoch": 0.3024061205550148,
      "grad_norm": 0.575445294380188,
      "learning_rate": 4.663851671744786e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 10755808,
      "step": 6680
    },
    {
      "epoch": 0.3026324724416578,
      "grad_norm": 0.732423722743988,
      "learning_rate": 4.6633598077184815e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 10763680,
      "step": 6685
    },
    {
      "epoch": 0.3028588243283008,
      "grad_norm": 0.6116673350334167,
      "learning_rate": 4.662867610074908e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 10771328,
      "step": 6690
    },
    {
      "epoch": 0.30308517621494374,
      "grad_norm": 0.8788788318634033,
      "learning_rate": 4.6623750788899696e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 10779232,
      "step": 6695
    },
    {
      "epoch": 0.30331152810158674,
      "grad_norm": 1.0044375658035278,
      "learning_rate": 4.6618822142396195e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 10787392,
      "step": 6700
    },
    {
      "epoch": 0.3035378799882297,
      "grad_norm": 0.4672558009624481,
      "learning_rate": 4.661389016199864e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 10794848,
      "step": 6705
    },
    {
      "epoch": 0.30376423187487267,
      "grad_norm": 0.5934774279594421,
      "learning_rate": 4.660895484846761e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 10803360,
      "step": 6710
    },
    {
      "epoch": 0.30399058376151566,
      "grad_norm": 0.871537446975708,
      "learning_rate": 4.660401620256418e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 10811616,
      "step": 6715
    },
    {
      "epoch": 0.3042169356481586,
      "grad_norm": 0.40958812832832336,
      "learning_rate": 4.659907422504997e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 10819520,
      "step": 6720
    },
    {
      "epoch": 0.3044432875348016,
      "grad_norm": 1.143652319908142,
      "learning_rate": 4.6594128916687074e-05,
      "loss": 0.2203,
      "num_input_tokens_seen": 10827424,
      "step": 6725
    },
    {
      "epoch": 0.3046696394214446,
      "grad_norm": 1.4034185409545898,
      "learning_rate": 4.658918027823813e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 10834752,
      "step": 6730
    },
    {
      "epoch": 0.30489599130808753,
      "grad_norm": 0.920634925365448,
      "learning_rate": 4.658422831046628e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 10842624,
      "step": 6735
    },
    {
      "epoch": 0.3051223431947305,
      "grad_norm": 0.3637601435184479,
      "learning_rate": 4.657927301413518e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 10850560,
      "step": 6740
    },
    {
      "epoch": 0.3053486950813735,
      "grad_norm": 0.3833715319633484,
      "learning_rate": 4.657431439000901e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 10858720,
      "step": 6745
    },
    {
      "epoch": 0.30557504696801646,
      "grad_norm": 0.4470384418964386,
      "learning_rate": 4.656935243885243e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 10866656,
      "step": 6750
    },
    {
      "epoch": 0.30580139885465946,
      "grad_norm": 0.35819000005722046,
      "learning_rate": 4.656438716143066e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 10875072,
      "step": 6755
    },
    {
      "epoch": 0.30602775074130245,
      "grad_norm": 0.66253662109375,
      "learning_rate": 4.6559418558509384e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 10883264,
      "step": 6760
    },
    {
      "epoch": 0.3062541026279454,
      "grad_norm": 0.7156515717506409,
      "learning_rate": 4.6554446630854833e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 10890880,
      "step": 6765
    },
    {
      "epoch": 0.3064804545145884,
      "grad_norm": 0.5916875004768372,
      "learning_rate": 4.654947137923374e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 10898624,
      "step": 6770
    },
    {
      "epoch": 0.3067068064012314,
      "grad_norm": 0.4661296606063843,
      "learning_rate": 4.654449280441335e-05,
      "loss": 0.1916,
      "num_input_tokens_seen": 10907008,
      "step": 6775
    },
    {
      "epoch": 0.3069331582878743,
      "grad_norm": 0.5141115784645081,
      "learning_rate": 4.653951090716143e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 10915328,
      "step": 6780
    },
    {
      "epoch": 0.3071595101745173,
      "grad_norm": 0.37959685921669006,
      "learning_rate": 4.653452568824625e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 10923424,
      "step": 6785
    },
    {
      "epoch": 0.30738586206116025,
      "grad_norm": 0.8910111784934998,
      "learning_rate": 4.6529537148436585e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 10931040,
      "step": 6790
    },
    {
      "epoch": 0.30761221394780325,
      "grad_norm": 0.7911384105682373,
      "learning_rate": 4.6524545288501734e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 10938784,
      "step": 6795
    },
    {
      "epoch": 0.30783856583444624,
      "grad_norm": 0.36096587777137756,
      "learning_rate": 4.6519550109211506e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 10947264,
      "step": 6800
    },
    {
      "epoch": 0.30783856583444624,
      "eval_loss": 0.16818967461585999,
      "eval_runtime": 404.1502,
      "eval_samples_per_second": 97.169,
      "eval_steps_per_second": 24.293,
      "num_input_tokens_seen": 10947264,
      "step": 6800
    },
    {
      "epoch": 0.3080649177210892,
      "grad_norm": 0.5204042196273804,
      "learning_rate": 4.651455161133622e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 10955232,
      "step": 6805
    },
    {
      "epoch": 0.3082912696077322,
      "grad_norm": 0.47061657905578613,
      "learning_rate": 4.6509549795646704e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 10963136,
      "step": 6810
    },
    {
      "epoch": 0.30851762149437517,
      "grad_norm": 0.30695250630378723,
      "learning_rate": 4.6504544662914306e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 10971008,
      "step": 6815
    },
    {
      "epoch": 0.3087439733810181,
      "grad_norm": 0.7042829990386963,
      "learning_rate": 4.6499536213910876e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 10978656,
      "step": 6820
    },
    {
      "epoch": 0.3089703252676611,
      "grad_norm": 0.8781282305717468,
      "learning_rate": 4.6494524449408786e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 10986240,
      "step": 6825
    },
    {
      "epoch": 0.3091966771543041,
      "grad_norm": 1.3443342447280884,
      "learning_rate": 4.6489509370180903e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 10993952,
      "step": 6830
    },
    {
      "epoch": 0.30942302904094704,
      "grad_norm": 0.5378208756446838,
      "learning_rate": 4.648449097700063e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 11001600,
      "step": 6835
    },
    {
      "epoch": 0.30964938092759003,
      "grad_norm": 1.0858154296875,
      "learning_rate": 4.647946927064185e-05,
      "loss": 0.2036,
      "num_input_tokens_seen": 11009856,
      "step": 6840
    },
    {
      "epoch": 0.30987573281423303,
      "grad_norm": 0.6787390112876892,
      "learning_rate": 4.647444425187898e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 11018080,
      "step": 6845
    },
    {
      "epoch": 0.31010208470087597,
      "grad_norm": 0.6000856161117554,
      "learning_rate": 4.646941592148695e-05,
      "loss": 0.2181,
      "num_input_tokens_seen": 11026016,
      "step": 6850
    },
    {
      "epoch": 0.31032843658751896,
      "grad_norm": 0.7504081130027771,
      "learning_rate": 4.646438428024117e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 11034272,
      "step": 6855
    },
    {
      "epoch": 0.31055478847416196,
      "grad_norm": 0.5958604216575623,
      "learning_rate": 4.64593493289176e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 11042336,
      "step": 6860
    },
    {
      "epoch": 0.3107811403608049,
      "grad_norm": 0.5810171365737915,
      "learning_rate": 4.64543110682927e-05,
      "loss": 0.2105,
      "num_input_tokens_seen": 11050304,
      "step": 6865
    },
    {
      "epoch": 0.3110074922474479,
      "grad_norm": 0.47143566608428955,
      "learning_rate": 4.644926949914341e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 11058432,
      "step": 6870
    },
    {
      "epoch": 0.31123384413409083,
      "grad_norm": 1.0958338975906372,
      "learning_rate": 4.644422462224722e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 11068128,
      "step": 6875
    },
    {
      "epoch": 0.3114601960207338,
      "grad_norm": 0.6900694370269775,
      "learning_rate": 4.643917643838211e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 11075744,
      "step": 6880
    },
    {
      "epoch": 0.3116865479073768,
      "grad_norm": 0.6808593273162842,
      "learning_rate": 4.6434124948326564e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 11083616,
      "step": 6885
    },
    {
      "epoch": 0.31191289979401976,
      "grad_norm": 0.3248114287853241,
      "learning_rate": 4.6429070152859594e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 11091904,
      "step": 6890
    },
    {
      "epoch": 0.31213925168066275,
      "grad_norm": 0.6489729881286621,
      "learning_rate": 4.6424012052760714e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 11100512,
      "step": 6895
    },
    {
      "epoch": 0.31236560356730575,
      "grad_norm": 0.518232524394989,
      "learning_rate": 4.6418950648809945e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 11108352,
      "step": 6900
    },
    {
      "epoch": 0.3125919554539487,
      "grad_norm": 0.8558483719825745,
      "learning_rate": 4.641388594178782e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 11116064,
      "step": 6905
    },
    {
      "epoch": 0.3128183073405917,
      "grad_norm": 0.7132518291473389,
      "learning_rate": 4.640881793247538e-05,
      "loss": 0.2143,
      "num_input_tokens_seen": 11124064,
      "step": 6910
    },
    {
      "epoch": 0.3130446592272347,
      "grad_norm": 0.566691517829895,
      "learning_rate": 4.6403746621654173e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 11131872,
      "step": 6915
    },
    {
      "epoch": 0.3132710111138776,
      "grad_norm": 0.4138906002044678,
      "learning_rate": 4.639867201010626e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 11140224,
      "step": 6920
    },
    {
      "epoch": 0.3134973630005206,
      "grad_norm": 0.32656827569007874,
      "learning_rate": 4.6393594098614204e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 11148512,
      "step": 6925
    },
    {
      "epoch": 0.3137237148871636,
      "grad_norm": 0.671999454498291,
      "learning_rate": 4.63885128879611e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 11156576,
      "step": 6930
    },
    {
      "epoch": 0.31395006677380655,
      "grad_norm": 0.7352738380432129,
      "learning_rate": 4.638342837893052e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 11164064,
      "step": 6935
    },
    {
      "epoch": 0.31417641866044954,
      "grad_norm": 0.6380496025085449,
      "learning_rate": 4.6378340572306565e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 11171968,
      "step": 6940
    },
    {
      "epoch": 0.31440277054709254,
      "grad_norm": 0.5656114220619202,
      "learning_rate": 4.6373249468873833e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 11179744,
      "step": 6945
    },
    {
      "epoch": 0.3146291224337355,
      "grad_norm": 0.37944045662879944,
      "learning_rate": 4.636815506941744e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 11187392,
      "step": 6950
    },
    {
      "epoch": 0.31485547432037847,
      "grad_norm": 0.4730021357536316,
      "learning_rate": 4.6363057374723004e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 11195040,
      "step": 6955
    },
    {
      "epoch": 0.3150818262070214,
      "grad_norm": 0.8740838170051575,
      "learning_rate": 4.635795638557666e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 11203840,
      "step": 6960
    },
    {
      "epoch": 0.3153081780936644,
      "grad_norm": 0.5600438714027405,
      "learning_rate": 4.635285210276504e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 11212096,
      "step": 6965
    },
    {
      "epoch": 0.3155345299803074,
      "grad_norm": 0.7403307557106018,
      "learning_rate": 4.6347744527075295e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 11220512,
      "step": 6970
    },
    {
      "epoch": 0.31576088186695034,
      "grad_norm": 0.43102309107780457,
      "learning_rate": 4.634263365929506e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 11228704,
      "step": 6975
    },
    {
      "epoch": 0.31598723375359333,
      "grad_norm": 0.810741126537323,
      "learning_rate": 4.6337519500212515e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 11237056,
      "step": 6980
    },
    {
      "epoch": 0.3162135856402363,
      "grad_norm": 0.5619869828224182,
      "learning_rate": 4.633240205061632e-05,
      "loss": 0.1909,
      "num_input_tokens_seen": 11245888,
      "step": 6985
    },
    {
      "epoch": 0.31643993752687927,
      "grad_norm": 0.9157131910324097,
      "learning_rate": 4.632728131129565e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 11253408,
      "step": 6990
    },
    {
      "epoch": 0.31666628941352226,
      "grad_norm": 0.726395845413208,
      "learning_rate": 4.632215728304018e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 11260992,
      "step": 6995
    },
    {
      "epoch": 0.31689264130016526,
      "grad_norm": 0.4050813615322113,
      "learning_rate": 4.63170299666401e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 11268736,
      "step": 7000
    },
    {
      "epoch": 0.31689264130016526,
      "eval_loss": 0.1694176346063614,
      "eval_runtime": 404.8495,
      "eval_samples_per_second": 97.001,
      "eval_steps_per_second": 24.251,
      "num_input_tokens_seen": 11268736,
      "step": 7000
    },
    {
      "epoch": 0.3171189931868082,
      "grad_norm": 0.44187483191490173,
      "learning_rate": 4.631189936288612e-05,
      "loss": 0.1766,
      "num_input_tokens_seen": 11276896,
      "step": 7005
    },
    {
      "epoch": 0.3173453450734512,
      "grad_norm": 0.5022150278091431,
      "learning_rate": 4.630676547256944e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 11284608,
      "step": 7010
    },
    {
      "epoch": 0.3175716969600942,
      "grad_norm": 0.5140979290008545,
      "learning_rate": 4.630162829648176e-05,
      "loss": 0.2057,
      "num_input_tokens_seen": 11292960,
      "step": 7015
    },
    {
      "epoch": 0.3177980488467371,
      "grad_norm": 0.636663019657135,
      "learning_rate": 4.629648783541531e-05,
      "loss": 0.2043,
      "num_input_tokens_seen": 11301312,
      "step": 7020
    },
    {
      "epoch": 0.3180244007333801,
      "grad_norm": 0.7183420658111572,
      "learning_rate": 4.6291344090162804e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 11309696,
      "step": 7025
    },
    {
      "epoch": 0.3182507526200231,
      "grad_norm": 0.41316279768943787,
      "learning_rate": 4.628619706151748e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 11318048,
      "step": 7030
    },
    {
      "epoch": 0.31847710450666605,
      "grad_norm": 0.9588456153869629,
      "learning_rate": 4.628104675027306e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 11326336,
      "step": 7035
    },
    {
      "epoch": 0.31870345639330905,
      "grad_norm": 1.1557472944259644,
      "learning_rate": 4.6275893157223805e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 11334752,
      "step": 7040
    },
    {
      "epoch": 0.318929808279952,
      "grad_norm": 0.44443637132644653,
      "learning_rate": 4.627073628316445e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 11342944,
      "step": 7045
    },
    {
      "epoch": 0.319156160166595,
      "grad_norm": 0.4286791980266571,
      "learning_rate": 4.626557612889026e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 11350656,
      "step": 7050
    },
    {
      "epoch": 0.319382512053238,
      "grad_norm": 0.5075589418411255,
      "learning_rate": 4.626041269519699e-05,
      "loss": 0.1902,
      "num_input_tokens_seen": 11358240,
      "step": 7055
    },
    {
      "epoch": 0.3196088639398809,
      "grad_norm": 0.4771583080291748,
      "learning_rate": 4.6255245982880905e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 11366304,
      "step": 7060
    },
    {
      "epoch": 0.3198352158265239,
      "grad_norm": 0.9419308304786682,
      "learning_rate": 4.625007599273879e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 11374304,
      "step": 7065
    },
    {
      "epoch": 0.3200615677131669,
      "grad_norm": 0.8608560562133789,
      "learning_rate": 4.6244902725567895e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 11382208,
      "step": 7070
    },
    {
      "epoch": 0.32028791959980984,
      "grad_norm": 0.4156840741634369,
      "learning_rate": 4.6239726182166024e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 11389472,
      "step": 7075
    },
    {
      "epoch": 0.32051427148645284,
      "grad_norm": 0.4992401897907257,
      "learning_rate": 4.623454636333147e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 11396768,
      "step": 7080
    },
    {
      "epoch": 0.32074062337309583,
      "grad_norm": 1.0478065013885498,
      "learning_rate": 4.622936326986301e-05,
      "loss": 0.1778,
      "num_input_tokens_seen": 11404896,
      "step": 7085
    },
    {
      "epoch": 0.3209669752597388,
      "grad_norm": 0.6620486974716187,
      "learning_rate": 4.6224176902559946e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 11412800,
      "step": 7090
    },
    {
      "epoch": 0.32119332714638177,
      "grad_norm": 0.6075605750083923,
      "learning_rate": 4.621898726222209e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 11421568,
      "step": 7095
    },
    {
      "epoch": 0.32141967903302476,
      "grad_norm": 1.2074241638183594,
      "learning_rate": 4.6213794349649744e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 11430112,
      "step": 7100
    },
    {
      "epoch": 0.3216460309196677,
      "grad_norm": 0.8320309519767761,
      "learning_rate": 4.6208598165643715e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 11437920,
      "step": 7105
    },
    {
      "epoch": 0.3218723828063107,
      "grad_norm": 0.5352150797843933,
      "learning_rate": 4.620339871100533e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 11445728,
      "step": 7110
    },
    {
      "epoch": 0.3220987346929537,
      "grad_norm": 0.9669299721717834,
      "learning_rate": 4.6198195986536394e-05,
      "loss": 0.2074,
      "num_input_tokens_seen": 11453376,
      "step": 7115
    },
    {
      "epoch": 0.32232508657959663,
      "grad_norm": 0.7897709012031555,
      "learning_rate": 4.619298999303926e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 11461632,
      "step": 7120
    },
    {
      "epoch": 0.3225514384662396,
      "grad_norm": 0.5529821515083313,
      "learning_rate": 4.618778073131673e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 11469536,
      "step": 7125
    },
    {
      "epoch": 0.32277779035288257,
      "grad_norm": 0.39318758249282837,
      "learning_rate": 4.618256820217215e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 11477088,
      "step": 7130
    },
    {
      "epoch": 0.32300414223952556,
      "grad_norm": 0.7024019956588745,
      "learning_rate": 4.617735240640936e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 11485632,
      "step": 7135
    },
    {
      "epoch": 0.32323049412616855,
      "grad_norm": 0.5379593372344971,
      "learning_rate": 4.6172133344832705e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 11493824,
      "step": 7140
    },
    {
      "epoch": 0.3234568460128115,
      "grad_norm": 0.8205090165138245,
      "learning_rate": 4.6166911018247004e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 11502400,
      "step": 7145
    },
    {
      "epoch": 0.3236831978994545,
      "grad_norm": 1.2305108308792114,
      "learning_rate": 4.616168542745764e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 11510048,
      "step": 7150
    },
    {
      "epoch": 0.3239095497860975,
      "grad_norm": 0.9324538111686707,
      "learning_rate": 4.6156456573270446e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 11518144,
      "step": 7155
    },
    {
      "epoch": 0.3241359016727404,
      "grad_norm": 0.5122026801109314,
      "learning_rate": 4.615122445649177e-05,
      "loss": 0.2005,
      "num_input_tokens_seen": 11527392,
      "step": 7160
    },
    {
      "epoch": 0.3243622535593834,
      "grad_norm": 0.30904141068458557,
      "learning_rate": 4.6145989077928486e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 11535776,
      "step": 7165
    },
    {
      "epoch": 0.3245886054460264,
      "grad_norm": 1.0865312814712524,
      "learning_rate": 4.6140750438387953e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 11544096,
      "step": 7170
    },
    {
      "epoch": 0.32481495733266935,
      "grad_norm": 0.5738177299499512,
      "learning_rate": 4.613550853867803e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 11553920,
      "step": 7175
    },
    {
      "epoch": 0.32504130921931235,
      "grad_norm": 0.5848453044891357,
      "learning_rate": 4.613026337960708e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 11561888,
      "step": 7180
    },
    {
      "epoch": 0.32526766110595534,
      "grad_norm": 0.9056294560432434,
      "learning_rate": 4.612501496198398e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 11570016,
      "step": 7185
    },
    {
      "epoch": 0.3254940129925983,
      "grad_norm": 0.5505159497261047,
      "learning_rate": 4.61197632866181e-05,
      "loss": 0.1881,
      "num_input_tokens_seen": 11577664,
      "step": 7190
    },
    {
      "epoch": 0.3257203648792413,
      "grad_norm": 0.461041122674942,
      "learning_rate": 4.611450835431931e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 11585568,
      "step": 7195
    },
    {
      "epoch": 0.32594671676588427,
      "grad_norm": 0.5943635702133179,
      "learning_rate": 4.6109250165898e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 11593952,
      "step": 7200
    },
    {
      "epoch": 0.32594671676588427,
      "eval_loss": 0.166753888130188,
      "eval_runtime": 404.289,
      "eval_samples_per_second": 97.136,
      "eval_steps_per_second": 24.285,
      "num_input_tokens_seen": 11593952,
      "step": 7200
    },
    {
      "epoch": 0.3261730686525272,
      "grad_norm": 1.0611923933029175,
      "learning_rate": 4.610398872216503e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 11601856,
      "step": 7205
    },
    {
      "epoch": 0.3263994205391702,
      "grad_norm": 0.8033203482627869,
      "learning_rate": 4.6098724023931796e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 11609440,
      "step": 7210
    },
    {
      "epoch": 0.32662577242581314,
      "grad_norm": 0.7971504926681519,
      "learning_rate": 4.609345607201017e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 11617472,
      "step": 7215
    },
    {
      "epoch": 0.32685212431245614,
      "grad_norm": 0.3565385341644287,
      "learning_rate": 4.608818486721254e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 11625216,
      "step": 7220
    },
    {
      "epoch": 0.32707847619909913,
      "grad_norm": 0.3605220913887024,
      "learning_rate": 4.608291041035179e-05,
      "loss": 0.214,
      "num_input_tokens_seen": 11633376,
      "step": 7225
    },
    {
      "epoch": 0.32730482808574207,
      "grad_norm": 0.5685427188873291,
      "learning_rate": 4.607763270224132e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 11641312,
      "step": 7230
    },
    {
      "epoch": 0.32753117997238507,
      "grad_norm": 0.6977685689926147,
      "learning_rate": 4.6072351743695e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 11648928,
      "step": 7235
    },
    {
      "epoch": 0.32775753185902806,
      "grad_norm": 0.596335232257843,
      "learning_rate": 4.606706753552723e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 11656672,
      "step": 7240
    },
    {
      "epoch": 0.327983883745671,
      "grad_norm": 0.2724129557609558,
      "learning_rate": 4.6061780078552906e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 11664704,
      "step": 7245
    },
    {
      "epoch": 0.328210235632314,
      "grad_norm": 0.348065584897995,
      "learning_rate": 4.605648937358742e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 11672864,
      "step": 7250
    },
    {
      "epoch": 0.328436587518957,
      "grad_norm": 0.6075959801673889,
      "learning_rate": 4.605119542144665e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 11681216,
      "step": 7255
    },
    {
      "epoch": 0.32866293940559993,
      "grad_norm": 1.0969021320343018,
      "learning_rate": 4.604589822294701e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 11689184,
      "step": 7260
    },
    {
      "epoch": 0.3288892912922429,
      "grad_norm": 0.6182344555854797,
      "learning_rate": 4.604059777890537e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 11697344,
      "step": 7265
    },
    {
      "epoch": 0.3291156431788859,
      "grad_norm": 0.49549543857574463,
      "learning_rate": 4.6035294090139145e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 11705024,
      "step": 7270
    },
    {
      "epoch": 0.32934199506552886,
      "grad_norm": 0.8942679762840271,
      "learning_rate": 4.6029987157466226e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 11712800,
      "step": 7275
    },
    {
      "epoch": 0.32956834695217185,
      "grad_norm": 0.8047173023223877,
      "learning_rate": 4.602467698170502e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 11720800,
      "step": 7280
    },
    {
      "epoch": 0.32979469883881485,
      "grad_norm": 0.6315629482269287,
      "learning_rate": 4.601936356367439e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 11728448,
      "step": 7285
    },
    {
      "epoch": 0.3300210507254578,
      "grad_norm": 0.9246299266815186,
      "learning_rate": 4.601404690419377e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 11736800,
      "step": 7290
    },
    {
      "epoch": 0.3302474026121008,
      "grad_norm": 0.9321327209472656,
      "learning_rate": 4.600872700408303e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 11744256,
      "step": 7295
    },
    {
      "epoch": 0.3304737544987437,
      "grad_norm": 0.5005162358283997,
      "learning_rate": 4.600340386416258e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 11752640,
      "step": 7300
    },
    {
      "epoch": 0.3307001063853867,
      "grad_norm": 0.8280841112136841,
      "learning_rate": 4.5998077485253296e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 11760640,
      "step": 7305
    },
    {
      "epoch": 0.3309264582720297,
      "grad_norm": 0.933949887752533,
      "learning_rate": 4.59927478681766e-05,
      "loss": 0.19,
      "num_input_tokens_seen": 11768640,
      "step": 7310
    },
    {
      "epoch": 0.33115281015867265,
      "grad_norm": 0.8899503946304321,
      "learning_rate": 4.5987415013754366e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 11776896,
      "step": 7315
    },
    {
      "epoch": 0.33137916204531565,
      "grad_norm": 0.7487133145332336,
      "learning_rate": 4.598207892280899e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 11784832,
      "step": 7320
    },
    {
      "epoch": 0.33160551393195864,
      "grad_norm": 0.614334762096405,
      "learning_rate": 4.597673959616337e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 11793120,
      "step": 7325
    },
    {
      "epoch": 0.3318318658186016,
      "grad_norm": 0.40729808807373047,
      "learning_rate": 4.597139703464089e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 11801024,
      "step": 7330
    },
    {
      "epoch": 0.3320582177052446,
      "grad_norm": 0.6614832282066345,
      "learning_rate": 4.596605123906545e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 11809280,
      "step": 7335
    },
    {
      "epoch": 0.33228456959188757,
      "grad_norm": 0.7652495503425598,
      "learning_rate": 4.596070221026143e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 11817312,
      "step": 7340
    },
    {
      "epoch": 0.3325109214785305,
      "grad_norm": 0.33030977845191956,
      "learning_rate": 4.595534994905372e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 11825216,
      "step": 7345
    },
    {
      "epoch": 0.3327372733651735,
      "grad_norm": 0.7719264030456543,
      "learning_rate": 4.594999445626771e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 11833184,
      "step": 7350
    },
    {
      "epoch": 0.3329636252518165,
      "grad_norm": 1.3716813325881958,
      "learning_rate": 4.5944635732729276e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 11841536,
      "step": 7355
    },
    {
      "epoch": 0.33318997713845944,
      "grad_norm": 1.143041729927063,
      "learning_rate": 4.5939273779264804e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 11849536,
      "step": 7360
    },
    {
      "epoch": 0.33341632902510243,
      "grad_norm": 0.5010690689086914,
      "learning_rate": 4.593390859670118e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 11857408,
      "step": 7365
    },
    {
      "epoch": 0.3336426809117454,
      "grad_norm": 0.8762469291687012,
      "learning_rate": 4.5928540185865776e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 11865472,
      "step": 7370
    },
    {
      "epoch": 0.33386903279838837,
      "grad_norm": 0.8676897287368774,
      "learning_rate": 4.592316854758648e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 11872960,
      "step": 7375
    },
    {
      "epoch": 0.33409538468503136,
      "grad_norm": 0.6975151896476746,
      "learning_rate": 4.5917793682691646e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 11880928,
      "step": 7380
    },
    {
      "epoch": 0.3343217365716743,
      "grad_norm": 0.510184109210968,
      "learning_rate": 4.5912415592010164e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 11888800,
      "step": 7385
    },
    {
      "epoch": 0.3345480884583173,
      "grad_norm": 0.49989965558052063,
      "learning_rate": 4.5907034276371386e-05,
      "loss": 0.1866,
      "num_input_tokens_seen": 11896992,
      "step": 7390
    },
    {
      "epoch": 0.3347744403449603,
      "grad_norm": 0.7794310450553894,
      "learning_rate": 4.5901649736605196e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 11904864,
      "step": 7395
    },
    {
      "epoch": 0.33500079223160323,
      "grad_norm": 0.6954095959663391,
      "learning_rate": 4.589626197354195e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 11912448,
      "step": 7400
    },
    {
      "epoch": 0.33500079223160323,
      "eval_loss": 0.1659231185913086,
      "eval_runtime": 403.8473,
      "eval_samples_per_second": 97.242,
      "eval_steps_per_second": 24.311,
      "num_input_tokens_seen": 11912448,
      "step": 7400
    },
    {
      "epoch": 0.3352271441182462,
      "grad_norm": 0.457746684551239,
      "learning_rate": 4.5890870988012504e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 11920224,
      "step": 7405
    },
    {
      "epoch": 0.3354534960048892,
      "grad_norm": 1.1793087720870972,
      "learning_rate": 4.5885476780848226e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 11928160,
      "step": 7410
    },
    {
      "epoch": 0.33567984789153216,
      "grad_norm": 0.9647333025932312,
      "learning_rate": 4.5880079352880964e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 11936128,
      "step": 7415
    },
    {
      "epoch": 0.33590619977817515,
      "grad_norm": 1.1561501026153564,
      "learning_rate": 4.5874678704943065e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 11943936,
      "step": 7420
    },
    {
      "epoch": 0.33613255166481815,
      "grad_norm": 0.33103275299072266,
      "learning_rate": 4.5869274837867394e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 11952352,
      "step": 7425
    },
    {
      "epoch": 0.3363589035514611,
      "grad_norm": 0.4168832004070282,
      "learning_rate": 4.5863867752487275e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 11960032,
      "step": 7430
    },
    {
      "epoch": 0.3365852554381041,
      "grad_norm": 0.49540144205093384,
      "learning_rate": 4.5858457449636554e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 11968000,
      "step": 7435
    },
    {
      "epoch": 0.3368116073247471,
      "grad_norm": 0.6393265724182129,
      "learning_rate": 4.5853043930149574e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 11976032,
      "step": 7440
    },
    {
      "epoch": 0.33703795921139,
      "grad_norm": 1.1845272779464722,
      "learning_rate": 4.584762719486117e-05,
      "loss": 0.199,
      "num_input_tokens_seen": 11983552,
      "step": 7445
    },
    {
      "epoch": 0.337264311098033,
      "grad_norm": 1.1325068473815918,
      "learning_rate": 4.584220724460665e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 11991104,
      "step": 7450
    },
    {
      "epoch": 0.337490662984676,
      "grad_norm": 1.1979738473892212,
      "learning_rate": 4.5836784080221865e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 11999712,
      "step": 7455
    },
    {
      "epoch": 0.33771701487131894,
      "grad_norm": 0.6651762127876282,
      "learning_rate": 4.583135770254312e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 12008064,
      "step": 7460
    },
    {
      "epoch": 0.33794336675796194,
      "grad_norm": 0.7624304890632629,
      "learning_rate": 4.5825928112407236e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 12015616,
      "step": 7465
    },
    {
      "epoch": 0.3381697186446049,
      "grad_norm": 1.0213810205459595,
      "learning_rate": 4.582049531065152e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 12023584,
      "step": 7470
    },
    {
      "epoch": 0.3383960705312479,
      "grad_norm": 0.9436916708946228,
      "learning_rate": 4.5815059298113783e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 12031360,
      "step": 7475
    },
    {
      "epoch": 0.33862242241789087,
      "grad_norm": 0.4618322253227234,
      "learning_rate": 4.580962007563232e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 12040032,
      "step": 7480
    },
    {
      "epoch": 0.3388487743045338,
      "grad_norm": 0.8350211381912231,
      "learning_rate": 4.5804177644045935e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 12047808,
      "step": 7485
    },
    {
      "epoch": 0.3390751261911768,
      "grad_norm": 0.8734135627746582,
      "learning_rate": 4.579873200419391e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 12055552,
      "step": 7490
    },
    {
      "epoch": 0.3393014780778198,
      "grad_norm": 0.8304691314697266,
      "learning_rate": 4.5793283156916046e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 12063264,
      "step": 7495
    },
    {
      "epoch": 0.33952782996446274,
      "grad_norm": 0.912774384021759,
      "learning_rate": 4.578783110305261e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 12070976,
      "step": 7500
    },
    {
      "epoch": 0.33975418185110573,
      "grad_norm": 0.5129635334014893,
      "learning_rate": 4.578237584344438e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 12078880,
      "step": 7505
    },
    {
      "epoch": 0.3399805337377487,
      "grad_norm": 0.7061490416526794,
      "learning_rate": 4.577691737893263e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 12086336,
      "step": 7510
    },
    {
      "epoch": 0.34020688562439166,
      "grad_norm": 0.2798357903957367,
      "learning_rate": 4.577145571035912e-05,
      "loss": 0.1866,
      "num_input_tokens_seen": 12093920,
      "step": 7515
    },
    {
      "epoch": 0.34043323751103466,
      "grad_norm": 0.6640765070915222,
      "learning_rate": 4.576599083856611e-05,
      "loss": 0.193,
      "num_input_tokens_seen": 12102560,
      "step": 7520
    },
    {
      "epoch": 0.34065958939767765,
      "grad_norm": 0.5507873296737671,
      "learning_rate": 4.576052276439635e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 12110112,
      "step": 7525
    },
    {
      "epoch": 0.3408859412843206,
      "grad_norm": 0.9904450178146362,
      "learning_rate": 4.575505148869308e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 12117792,
      "step": 7530
    },
    {
      "epoch": 0.3411122931709636,
      "grad_norm": 1.3606847524642944,
      "learning_rate": 4.574957701230006e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 12125920,
      "step": 7535
    },
    {
      "epoch": 0.3413386450576066,
      "grad_norm": 0.6780964732170105,
      "learning_rate": 4.57440993360615e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 12133632,
      "step": 7540
    },
    {
      "epoch": 0.3415649969442495,
      "grad_norm": 1.225958228111267,
      "learning_rate": 4.5738618460822134e-05,
      "loss": 0.2171,
      "num_input_tokens_seen": 12141344,
      "step": 7545
    },
    {
      "epoch": 0.3417913488308925,
      "grad_norm": 0.6075130701065063,
      "learning_rate": 4.573313438742719e-05,
      "loss": 0.2041,
      "num_input_tokens_seen": 12149728,
      "step": 7550
    },
    {
      "epoch": 0.34201770071753546,
      "grad_norm": 0.5906866788864136,
      "learning_rate": 4.5727647116722374e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 12158080,
      "step": 7555
    },
    {
      "epoch": 0.34224405260417845,
      "grad_norm": 0.4909795820713043,
      "learning_rate": 4.5722156649553884e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 12166176,
      "step": 7560
    },
    {
      "epoch": 0.34247040449082145,
      "grad_norm": 0.43432682752609253,
      "learning_rate": 4.571666298676843e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 12174048,
      "step": 7565
    },
    {
      "epoch": 0.3426967563774644,
      "grad_norm": 0.8046930432319641,
      "learning_rate": 4.571116612921321e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 12181504,
      "step": 7570
    },
    {
      "epoch": 0.3429231082641074,
      "grad_norm": 0.4420482814311981,
      "learning_rate": 4.57056660777359e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 12189600,
      "step": 7575
    },
    {
      "epoch": 0.3431494601507504,
      "grad_norm": 0.5683296918869019,
      "learning_rate": 4.5700162833184666e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 12197536,
      "step": 7580
    },
    {
      "epoch": 0.3433758120373933,
      "grad_norm": 0.9515591859817505,
      "learning_rate": 4.5694656396408195e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 12205024,
      "step": 7585
    },
    {
      "epoch": 0.3436021639240363,
      "grad_norm": 0.9388334155082703,
      "learning_rate": 4.5689146768255646e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 12213536,
      "step": 7590
    },
    {
      "epoch": 0.3438285158106793,
      "grad_norm": 0.5964802503585815,
      "learning_rate": 4.568363394957667e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 12221792,
      "step": 7595
    },
    {
      "epoch": 0.34405486769732224,
      "grad_norm": 0.42673155665397644,
      "learning_rate": 4.567811794122141e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 12230144,
      "step": 7600
    },
    {
      "epoch": 0.34405486769732224,
      "eval_loss": 0.1651124209165573,
      "eval_runtime": 404.6561,
      "eval_samples_per_second": 97.048,
      "eval_steps_per_second": 24.263,
      "num_input_tokens_seen": 12230144,
      "step": 7600
    },
    {
      "epoch": 0.34428121958396524,
      "grad_norm": 0.40454667806625366,
      "learning_rate": 4.56725987440405e-05,
      "loss": 0.2049,
      "num_input_tokens_seen": 12238144,
      "step": 7605
    },
    {
      "epoch": 0.34450757147060823,
      "grad_norm": 0.6109864711761475,
      "learning_rate": 4.566707635888508e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 12246112,
      "step": 7610
    },
    {
      "epoch": 0.34473392335725117,
      "grad_norm": 0.7080358266830444,
      "learning_rate": 4.566155078660677e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 12254880,
      "step": 7615
    },
    {
      "epoch": 0.34496027524389417,
      "grad_norm": 0.6131980419158936,
      "learning_rate": 4.565602202805768e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 12263200,
      "step": 7620
    },
    {
      "epoch": 0.34518662713053716,
      "grad_norm": 0.8297763466835022,
      "learning_rate": 4.56504900840904e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 12271264,
      "step": 7625
    },
    {
      "epoch": 0.3454129790171801,
      "grad_norm": 0.6021725535392761,
      "learning_rate": 4.564495495555805e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 12278432,
      "step": 7630
    },
    {
      "epoch": 0.3456393309038231,
      "grad_norm": 0.6006028056144714,
      "learning_rate": 4.5639416643314204e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 12286656,
      "step": 7635
    },
    {
      "epoch": 0.34586568279046603,
      "grad_norm": 0.6169829368591309,
      "learning_rate": 4.5633875148212946e-05,
      "loss": 0.1519,
      "num_input_tokens_seen": 12294016,
      "step": 7640
    },
    {
      "epoch": 0.34609203467710903,
      "grad_norm": 0.44341418147087097,
      "learning_rate": 4.562833047110883e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 12301888,
      "step": 7645
    },
    {
      "epoch": 0.346318386563752,
      "grad_norm": 0.3366917669773102,
      "learning_rate": 4.5622782612856923e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 12309632,
      "step": 7650
    },
    {
      "epoch": 0.34654473845039496,
      "grad_norm": 0.40101534128189087,
      "learning_rate": 4.561723157431278e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 12317696,
      "step": 7655
    },
    {
      "epoch": 0.34677109033703796,
      "grad_norm": 1.4267182350158691,
      "learning_rate": 4.5611677356332435e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 12325824,
      "step": 7660
    },
    {
      "epoch": 0.34699744222368095,
      "grad_norm": 1.2241389751434326,
      "learning_rate": 4.560611995977242e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 12333824,
      "step": 7665
    },
    {
      "epoch": 0.3472237941103239,
      "grad_norm": 0.8130103945732117,
      "learning_rate": 4.560055938548975e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 12342016,
      "step": 7670
    },
    {
      "epoch": 0.3474501459969669,
      "grad_norm": 0.6976045966148376,
      "learning_rate": 4.5594995634341944e-05,
      "loss": 0.212,
      "num_input_tokens_seen": 12350272,
      "step": 7675
    },
    {
      "epoch": 0.3476764978836099,
      "grad_norm": 0.7879504561424255,
      "learning_rate": 4.5589428707187e-05,
      "loss": 0.1236,
      "num_input_tokens_seen": 12358016,
      "step": 7680
    },
    {
      "epoch": 0.3479028497702528,
      "grad_norm": 0.4594006836414337,
      "learning_rate": 4.55838586048834e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 12366112,
      "step": 7685
    },
    {
      "epoch": 0.3481292016568958,
      "grad_norm": 0.9745253324508667,
      "learning_rate": 4.557828532829013e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 12374592,
      "step": 7690
    },
    {
      "epoch": 0.3483555535435388,
      "grad_norm": 0.7849210500717163,
      "learning_rate": 4.557270887826667e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 12382592,
      "step": 7695
    },
    {
      "epoch": 0.34858190543018175,
      "grad_norm": 1.5344045162200928,
      "learning_rate": 4.556712925567296e-05,
      "loss": 0.2108,
      "num_input_tokens_seen": 12390464,
      "step": 7700
    },
    {
      "epoch": 0.34880825731682474,
      "grad_norm": 0.8958185911178589,
      "learning_rate": 4.5561546461369454e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 12398656,
      "step": 7705
    },
    {
      "epoch": 0.34903460920346774,
      "grad_norm": 1.0117638111114502,
      "learning_rate": 4.55559604962171e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 12406240,
      "step": 7710
    },
    {
      "epoch": 0.3492609610901107,
      "grad_norm": 0.871150016784668,
      "learning_rate": 4.55503713610773e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 12414240,
      "step": 7715
    },
    {
      "epoch": 0.3494873129767537,
      "grad_norm": 0.4711974263191223,
      "learning_rate": 4.5544779056812e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 12422240,
      "step": 7720
    },
    {
      "epoch": 0.3497136648633966,
      "grad_norm": 1.3029345273971558,
      "learning_rate": 4.553918358428358e-05,
      "loss": 0.1961,
      "num_input_tokens_seen": 12430176,
      "step": 7725
    },
    {
      "epoch": 0.3499400167500396,
      "grad_norm": 1.4049901962280273,
      "learning_rate": 4.553358494435494e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 12438816,
      "step": 7730
    },
    {
      "epoch": 0.3501663686366826,
      "grad_norm": 0.8434315323829651,
      "learning_rate": 4.5527983137889464e-05,
      "loss": 0.2168,
      "num_input_tokens_seen": 12447936,
      "step": 7735
    },
    {
      "epoch": 0.35039272052332554,
      "grad_norm": 1.3577097654342651,
      "learning_rate": 4.5522378165751015e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 12456064,
      "step": 7740
    },
    {
      "epoch": 0.35061907240996854,
      "grad_norm": 0.8023062348365784,
      "learning_rate": 4.5516770028803954e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 12463936,
      "step": 7745
    },
    {
      "epoch": 0.35084542429661153,
      "grad_norm": 1.087349772453308,
      "learning_rate": 4.5511158727913116e-05,
      "loss": 0.1894,
      "num_input_tokens_seen": 12471648,
      "step": 7750
    },
    {
      "epoch": 0.35107177618325447,
      "grad_norm": 0.7941467761993408,
      "learning_rate": 4.5505544263943856e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 12479968,
      "step": 7755
    },
    {
      "epoch": 0.35129812806989746,
      "grad_norm": 0.7057943344116211,
      "learning_rate": 4.549992663776197e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 12487744,
      "step": 7760
    },
    {
      "epoch": 0.35152447995654046,
      "grad_norm": 0.9676797389984131,
      "learning_rate": 4.5494305850233786e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 12495744,
      "step": 7765
    },
    {
      "epoch": 0.3517508318431834,
      "grad_norm": 0.47146645188331604,
      "learning_rate": 4.5488681902226094e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 12503648,
      "step": 7770
    },
    {
      "epoch": 0.3519771837298264,
      "grad_norm": 0.9571061730384827,
      "learning_rate": 4.5483054794606174e-05,
      "loss": 0.1663,
      "num_input_tokens_seen": 12511872,
      "step": 7775
    },
    {
      "epoch": 0.3522035356164694,
      "grad_norm": 1.2267916202545166,
      "learning_rate": 4.547742452824179e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 12519264,
      "step": 7780
    },
    {
      "epoch": 0.3524298875031123,
      "grad_norm": 0.44545185565948486,
      "learning_rate": 4.5471791104001215e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 12527200,
      "step": 7785
    },
    {
      "epoch": 0.3526562393897553,
      "grad_norm": 0.5989987850189209,
      "learning_rate": 4.546615452275319e-05,
      "loss": 0.1928,
      "num_input_tokens_seen": 12535136,
      "step": 7790
    },
    {
      "epoch": 0.3528825912763983,
      "grad_norm": 0.5183851718902588,
      "learning_rate": 4.5460514785366944e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 12543232,
      "step": 7795
    },
    {
      "epoch": 0.35310894316304126,
      "grad_norm": 0.5850935578346252,
      "learning_rate": 4.545487189271219e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 12551712,
      "step": 7800
    },
    {
      "epoch": 0.35310894316304126,
      "eval_loss": 0.16528785228729248,
      "eval_runtime": 404.0022,
      "eval_samples_per_second": 97.205,
      "eval_steps_per_second": 24.302,
      "num_input_tokens_seen": 12551712,
      "step": 7800
    },
    {
      "epoch": 0.35333529504968425,
      "grad_norm": 0.7235561013221741,
      "learning_rate": 4.544922584565914e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 12560256,
      "step": 7805
    },
    {
      "epoch": 0.3535616469363272,
      "grad_norm": 0.9104962944984436,
      "learning_rate": 4.544357664507848e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 12568224,
      "step": 7810
    },
    {
      "epoch": 0.3537879988229702,
      "grad_norm": 0.3940437138080597,
      "learning_rate": 4.54379242918414e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 12576832,
      "step": 7815
    },
    {
      "epoch": 0.3540143507096132,
      "grad_norm": 1.4417448043823242,
      "learning_rate": 4.543226878681955e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 12584416,
      "step": 7820
    },
    {
      "epoch": 0.3542407025962561,
      "grad_norm": 0.5752724409103394,
      "learning_rate": 4.5426610130885087e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 12592032,
      "step": 7825
    },
    {
      "epoch": 0.3544670544828991,
      "grad_norm": 0.47283539175987244,
      "learning_rate": 4.542094832491064e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 12601024,
      "step": 7830
    },
    {
      "epoch": 0.3546934063695421,
      "grad_norm": 1.0582314729690552,
      "learning_rate": 4.541528336976934e-05,
      "loss": 0.1979,
      "num_input_tokens_seen": 12608800,
      "step": 7835
    },
    {
      "epoch": 0.35491975825618505,
      "grad_norm": 0.7480169534683228,
      "learning_rate": 4.540961526633479e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 12617184,
      "step": 7840
    },
    {
      "epoch": 0.35514611014282804,
      "grad_norm": 0.4611451327800751,
      "learning_rate": 4.540394401548108e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 12624928,
      "step": 7845
    },
    {
      "epoch": 0.35537246202947104,
      "grad_norm": 0.6254627704620361,
      "learning_rate": 4.539826961808279e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 12632640,
      "step": 7850
    },
    {
      "epoch": 0.355598813916114,
      "grad_norm": 0.6888511776924133,
      "learning_rate": 4.5392592075014994e-05,
      "loss": 0.2093,
      "num_input_tokens_seen": 12640672,
      "step": 7855
    },
    {
      "epoch": 0.35582516580275697,
      "grad_norm": 0.45979079604148865,
      "learning_rate": 4.538691138715322e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 12648864,
      "step": 7860
    },
    {
      "epoch": 0.35605151768939997,
      "grad_norm": 0.5245469212532043,
      "learning_rate": 4.5381227555373516e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 12656800,
      "step": 7865
    },
    {
      "epoch": 0.3562778695760429,
      "grad_norm": 1.1936384439468384,
      "learning_rate": 4.537554058055239e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 12664480,
      "step": 7870
    },
    {
      "epoch": 0.3565042214626859,
      "grad_norm": 0.680441677570343,
      "learning_rate": 4.5369850463566865e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 12672096,
      "step": 7875
    },
    {
      "epoch": 0.3567305733493289,
      "grad_norm": 0.4995081424713135,
      "learning_rate": 4.5364157205294404e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 12680320,
      "step": 7880
    },
    {
      "epoch": 0.35695692523597183,
      "grad_norm": 0.523920476436615,
      "learning_rate": 4.5358460806612996e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 12687840,
      "step": 7885
    },
    {
      "epoch": 0.35718327712261483,
      "grad_norm": 0.7866888046264648,
      "learning_rate": 4.535276126840109e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 12695648,
      "step": 7890
    },
    {
      "epoch": 0.35740962900925777,
      "grad_norm": 0.4152379035949707,
      "learning_rate": 4.5347058591537626e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 12703648,
      "step": 7895
    },
    {
      "epoch": 0.35763598089590076,
      "grad_norm": 1.1338801383972168,
      "learning_rate": 4.534135277690203e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 12710848,
      "step": 7900
    },
    {
      "epoch": 0.35786233278254376,
      "grad_norm": 0.9786465167999268,
      "learning_rate": 4.533564382537421e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 12718336,
      "step": 7905
    },
    {
      "epoch": 0.3580886846691867,
      "grad_norm": 0.5771664381027222,
      "learning_rate": 4.532993173783456e-05,
      "loss": 0.2016,
      "num_input_tokens_seen": 12726560,
      "step": 7910
    },
    {
      "epoch": 0.3583150365558297,
      "grad_norm": 0.6838205456733704,
      "learning_rate": 4.5324216515163954e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 12734592,
      "step": 7915
    },
    {
      "epoch": 0.3585413884424727,
      "grad_norm": 0.46023455262184143,
      "learning_rate": 4.531849815824375e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 12742016,
      "step": 7920
    },
    {
      "epoch": 0.3587677403291156,
      "grad_norm": 0.7524197101593018,
      "learning_rate": 4.5312776667955795e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 12749856,
      "step": 7925
    },
    {
      "epoch": 0.3589940922157586,
      "grad_norm": 0.6130293607711792,
      "learning_rate": 4.5307052045182405e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 12758016,
      "step": 7930
    },
    {
      "epoch": 0.3592204441024016,
      "grad_norm": 0.6713601350784302,
      "learning_rate": 4.53013242908064e-05,
      "loss": 0.195,
      "num_input_tokens_seen": 12765888,
      "step": 7935
    },
    {
      "epoch": 0.35944679598904455,
      "grad_norm": 0.4686376452445984,
      "learning_rate": 4.529559340571107e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 12774336,
      "step": 7940
    },
    {
      "epoch": 0.35967314787568755,
      "grad_norm": 0.5662498474121094,
      "learning_rate": 4.528985939078018e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 12782368,
      "step": 7945
    },
    {
      "epoch": 0.35989949976233054,
      "grad_norm": 1.4927175045013428,
      "learning_rate": 4.5284122246898e-05,
      "loss": 0.2144,
      "num_input_tokens_seen": 12789952,
      "step": 7950
    },
    {
      "epoch": 0.3601258516489735,
      "grad_norm": 1.3588145971298218,
      "learning_rate": 4.527838197494926e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 12797760,
      "step": 7955
    },
    {
      "epoch": 0.3603522035356165,
      "grad_norm": 0.9418195486068726,
      "learning_rate": 4.527263857581918e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 12805600,
      "step": 7960
    },
    {
      "epoch": 0.3605785554222594,
      "grad_norm": 0.6652377843856812,
      "learning_rate": 4.526689205039347e-05,
      "loss": 0.2007,
      "num_input_tokens_seen": 12813248,
      "step": 7965
    },
    {
      "epoch": 0.3608049073089024,
      "grad_norm": 0.449448823928833,
      "learning_rate": 4.5261142399558324e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 12821408,
      "step": 7970
    },
    {
      "epoch": 0.3610312591955454,
      "grad_norm": 0.6114409565925598,
      "learning_rate": 4.525538962420041e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 12828896,
      "step": 7975
    },
    {
      "epoch": 0.36125761108218835,
      "grad_norm": 0.317579060792923,
      "learning_rate": 4.524963372520685e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 12836480,
      "step": 7980
    },
    {
      "epoch": 0.36148396296883134,
      "grad_norm": 0.4753504991531372,
      "learning_rate": 4.524387470346531e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 12844992,
      "step": 7985
    },
    {
      "epoch": 0.36171031485547434,
      "grad_norm": 0.7278187274932861,
      "learning_rate": 4.5238112559863885e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 12853888,
      "step": 7990
    },
    {
      "epoch": 0.3619366667421173,
      "grad_norm": 0.582081139087677,
      "learning_rate": 4.5232347295291175e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 12861952,
      "step": 7995
    },
    {
      "epoch": 0.36216301862876027,
      "grad_norm": 0.5861408710479736,
      "learning_rate": 4.522657891063626e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 12870016,
      "step": 8000
    },
    {
      "epoch": 0.36216301862876027,
      "eval_loss": 0.16435988247394562,
      "eval_runtime": 404.2376,
      "eval_samples_per_second": 97.148,
      "eval_steps_per_second": 24.288,
      "num_input_tokens_seen": 12870016,
      "step": 8000
    },
    {
      "epoch": 0.36238937051540326,
      "grad_norm": 0.5688992142677307,
      "learning_rate": 4.52208074067887e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 12878080,
      "step": 8005
    },
    {
      "epoch": 0.3626157224020462,
      "grad_norm": 0.7231574654579163,
      "learning_rate": 4.5215032784638516e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 12886528,
      "step": 8010
    },
    {
      "epoch": 0.3628420742886892,
      "grad_norm": 0.7836295962333679,
      "learning_rate": 4.5209255045076245e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 12895136,
      "step": 8015
    },
    {
      "epoch": 0.3630684261753322,
      "grad_norm": 0.6423409581184387,
      "learning_rate": 4.5203474188992875e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 12903008,
      "step": 8020
    },
    {
      "epoch": 0.36329477806197513,
      "grad_norm": 0.7360284924507141,
      "learning_rate": 4.51976902172799e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 12911296,
      "step": 8025
    },
    {
      "epoch": 0.3635211299486181,
      "grad_norm": 0.8711978793144226,
      "learning_rate": 4.519190313082927e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 12919104,
      "step": 8030
    },
    {
      "epoch": 0.3637474818352611,
      "grad_norm": 0.6850529313087463,
      "learning_rate": 4.518611293053343e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 12926656,
      "step": 8035
    },
    {
      "epoch": 0.36397383372190406,
      "grad_norm": 0.5638952255249023,
      "learning_rate": 4.51803196172853e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 12935456,
      "step": 8040
    },
    {
      "epoch": 0.36420018560854706,
      "grad_norm": 0.32629263401031494,
      "learning_rate": 4.517452319197828e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 12943904,
      "step": 8045
    },
    {
      "epoch": 0.36442653749519,
      "grad_norm": 1.1126394271850586,
      "learning_rate": 4.5168723655506265e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 12952256,
      "step": 8050
    },
    {
      "epoch": 0.364652889381833,
      "grad_norm": 0.5329020023345947,
      "learning_rate": 4.51629210087636e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 12960416,
      "step": 8055
    },
    {
      "epoch": 0.364879241268476,
      "grad_norm": 0.7614567875862122,
      "learning_rate": 4.515711525264513e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 12968928,
      "step": 8060
    },
    {
      "epoch": 0.3651055931551189,
      "grad_norm": 0.7645530700683594,
      "learning_rate": 4.5151306388046175e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 12977120,
      "step": 8065
    },
    {
      "epoch": 0.3653319450417619,
      "grad_norm": 1.0111322402954102,
      "learning_rate": 4.514549441586255e-05,
      "loss": 0.191,
      "num_input_tokens_seen": 12985248,
      "step": 8070
    },
    {
      "epoch": 0.3655582969284049,
      "grad_norm": 0.5543989539146423,
      "learning_rate": 4.513967933699051e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 12992480,
      "step": 8075
    },
    {
      "epoch": 0.36578464881504785,
      "grad_norm": 0.360452264547348,
      "learning_rate": 4.513386115232684e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 13000768,
      "step": 8080
    },
    {
      "epoch": 0.36601100070169085,
      "grad_norm": 0.3933972716331482,
      "learning_rate": 4.5128039862768745e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 13008640,
      "step": 8085
    },
    {
      "epoch": 0.36623735258833384,
      "grad_norm": 0.5865612030029297,
      "learning_rate": 4.512221546921397e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 13016992,
      "step": 8090
    },
    {
      "epoch": 0.3664637044749768,
      "grad_norm": 0.8414588570594788,
      "learning_rate": 4.5116387972560694e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 13024672,
      "step": 8095
    },
    {
      "epoch": 0.3666900563616198,
      "grad_norm": 0.6174033880233765,
      "learning_rate": 4.511055737370759e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 13032960,
      "step": 8100
    },
    {
      "epoch": 0.36691640824826277,
      "grad_norm": 0.4568599462509155,
      "learning_rate": 4.510472367355383e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 13040992,
      "step": 8105
    },
    {
      "epoch": 0.3671427601349057,
      "grad_norm": 0.46074098348617554,
      "learning_rate": 4.509888687299901e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 13049824,
      "step": 8110
    },
    {
      "epoch": 0.3673691120215487,
      "grad_norm": 0.6091399788856506,
      "learning_rate": 4.5093046972943266e-05,
      "loss": 0.1985,
      "num_input_tokens_seen": 13057760,
      "step": 8115
    },
    {
      "epoch": 0.3675954639081917,
      "grad_norm": 0.8668208122253418,
      "learning_rate": 4.508720397428717e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 13065472,
      "step": 8120
    },
    {
      "epoch": 0.36782181579483464,
      "grad_norm": 0.32383230328559875,
      "learning_rate": 4.508135787793178e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 13074080,
      "step": 8125
    },
    {
      "epoch": 0.36804816768147763,
      "grad_norm": 0.6289477348327637,
      "learning_rate": 4.5075508684778664e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 13081920,
      "step": 8130
    },
    {
      "epoch": 0.3682745195681206,
      "grad_norm": 0.26804250478744507,
      "learning_rate": 4.506965639572982e-05,
      "loss": 0.1915,
      "num_input_tokens_seen": 13090208,
      "step": 8135
    },
    {
      "epoch": 0.36850087145476357,
      "grad_norm": 1.5664427280426025,
      "learning_rate": 4.506380101168774e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 13097952,
      "step": 8140
    },
    {
      "epoch": 0.36872722334140656,
      "grad_norm": 0.5726890563964844,
      "learning_rate": 4.505794253355542e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 13106208,
      "step": 8145
    },
    {
      "epoch": 0.3689535752280495,
      "grad_norm": 0.48195409774780273,
      "learning_rate": 4.5052080962236286e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 13114592,
      "step": 8150
    },
    {
      "epoch": 0.3691799271146925,
      "grad_norm": 1.2099124193191528,
      "learning_rate": 4.504621629863428e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 13122368,
      "step": 8155
    },
    {
      "epoch": 0.3694062790013355,
      "grad_norm": 0.5059807896614075,
      "learning_rate": 4.504034854365381e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 13130016,
      "step": 8160
    },
    {
      "epoch": 0.36963263088797843,
      "grad_norm": 0.7632249593734741,
      "learning_rate": 4.503447769819974e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 13137856,
      "step": 8165
    },
    {
      "epoch": 0.3698589827746214,
      "grad_norm": 1.320788860321045,
      "learning_rate": 4.502860376317745e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 13145312,
      "step": 8170
    },
    {
      "epoch": 0.3700853346612644,
      "grad_norm": 0.5704455971717834,
      "learning_rate": 4.502272673949276e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 13152960,
      "step": 8175
    },
    {
      "epoch": 0.37031168654790736,
      "grad_norm": 0.6472970843315125,
      "learning_rate": 4.501684662805199e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 13160832,
      "step": 8180
    },
    {
      "epoch": 0.37053803843455035,
      "grad_norm": 0.913513720035553,
      "learning_rate": 4.5010963429761924e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 13168384,
      "step": 8185
    },
    {
      "epoch": 0.37076439032119335,
      "grad_norm": 0.4347080588340759,
      "learning_rate": 4.500507714552982e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 13176384,
      "step": 8190
    },
    {
      "epoch": 0.3709907422078363,
      "grad_norm": 1.0563169717788696,
      "learning_rate": 4.499918777626342e-05,
      "loss": 0.2418,
      "num_input_tokens_seen": 13184512,
      "step": 8195
    },
    {
      "epoch": 0.3712170940944793,
      "grad_norm": 0.7639606595039368,
      "learning_rate": 4.499329532287093e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 13192096,
      "step": 8200
    },
    {
      "epoch": 0.3712170940944793,
      "eval_loss": 0.16332434117794037,
      "eval_runtime": 404.3772,
      "eval_samples_per_second": 97.115,
      "eval_steps_per_second": 24.279,
      "num_input_tokens_seen": 13192096,
      "step": 8200
    },
    {
      "epoch": 0.3714434459811223,
      "grad_norm": 0.74309241771698,
      "learning_rate": 4.4987399786261064e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 13199648,
      "step": 8205
    },
    {
      "epoch": 0.3716697978677652,
      "grad_norm": 0.6686946153640747,
      "learning_rate": 4.498150116734297e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 13207040,
      "step": 8210
    },
    {
      "epoch": 0.3718961497544082,
      "grad_norm": 0.32734331488609314,
      "learning_rate": 4.4975599467026294e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 13215616,
      "step": 8215
    },
    {
      "epoch": 0.37212250164105115,
      "grad_norm": 0.3705374002456665,
      "learning_rate": 4.496969468622114e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 13223232,
      "step": 8220
    },
    {
      "epoch": 0.37234885352769415,
      "grad_norm": 0.5486555099487305,
      "learning_rate": 4.496378682583813e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 13232192,
      "step": 8225
    },
    {
      "epoch": 0.37257520541433714,
      "grad_norm": 0.8857973217964172,
      "learning_rate": 4.495787588678829e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 13240704,
      "step": 8230
    },
    {
      "epoch": 0.3728015573009801,
      "grad_norm": 1.2672882080078125,
      "learning_rate": 4.4951961869983196e-05,
      "loss": 0.2048,
      "num_input_tokens_seen": 13248832,
      "step": 8235
    },
    {
      "epoch": 0.3730279091876231,
      "grad_norm": 0.7345110177993774,
      "learning_rate": 4.494604477633485e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 13256736,
      "step": 8240
    },
    {
      "epoch": 0.37325426107426607,
      "grad_norm": 0.4212612211704254,
      "learning_rate": 4.4940124606755734e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 13265088,
      "step": 8245
    },
    {
      "epoch": 0.373480612960909,
      "grad_norm": 0.5606656074523926,
      "learning_rate": 4.493420136215882e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 13272864,
      "step": 8250
    },
    {
      "epoch": 0.373706964847552,
      "grad_norm": 0.5141316056251526,
      "learning_rate": 4.492827504345756e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 13280576,
      "step": 8255
    },
    {
      "epoch": 0.373933316734195,
      "grad_norm": 0.59575355052948,
      "learning_rate": 4.492234565156584e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 13288352,
      "step": 8260
    },
    {
      "epoch": 0.37415966862083794,
      "grad_norm": 0.61909019947052,
      "learning_rate": 4.491641318739807e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 13296096,
      "step": 8265
    },
    {
      "epoch": 0.37438602050748093,
      "grad_norm": 0.7770200371742249,
      "learning_rate": 4.4910477651869096e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 13304608,
      "step": 8270
    },
    {
      "epoch": 0.3746123723941239,
      "grad_norm": 0.34693261981010437,
      "learning_rate": 4.4904539045894254e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 13312768,
      "step": 8275
    },
    {
      "epoch": 0.37483872428076687,
      "grad_norm": 0.8150938749313354,
      "learning_rate": 4.4898597370389364e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 13321120,
      "step": 8280
    },
    {
      "epoch": 0.37506507616740986,
      "grad_norm": 0.8606621623039246,
      "learning_rate": 4.489265262627069e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 13328672,
      "step": 8285
    },
    {
      "epoch": 0.37529142805405286,
      "grad_norm": 0.2926696538925171,
      "learning_rate": 4.488670481445499e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 13337728,
      "step": 8290
    },
    {
      "epoch": 0.3755177799406958,
      "grad_norm": 0.3825838267803192,
      "learning_rate": 4.488075393585951e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 13345888,
      "step": 8295
    },
    {
      "epoch": 0.3757441318273388,
      "grad_norm": 0.8197978734970093,
      "learning_rate": 4.487479999140193e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 13353600,
      "step": 8300
    },
    {
      "epoch": 0.37597048371398173,
      "grad_norm": 0.3726164698600769,
      "learning_rate": 4.4868842982000425e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 13361696,
      "step": 8305
    },
    {
      "epoch": 0.3761968356006247,
      "grad_norm": 0.9458532333374023,
      "learning_rate": 4.486288290857365e-05,
      "loss": 0.2088,
      "num_input_tokens_seen": 13369280,
      "step": 8310
    },
    {
      "epoch": 0.3764231874872677,
      "grad_norm": 0.809442400932312,
      "learning_rate": 4.4856919772040715e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 13377088,
      "step": 8315
    },
    {
      "epoch": 0.37664953937391066,
      "grad_norm": 0.37106579542160034,
      "learning_rate": 4.485095357332122e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 13385280,
      "step": 8320
    },
    {
      "epoch": 0.37687589126055365,
      "grad_norm": 1.3212072849273682,
      "learning_rate": 4.484498431333521e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 13393152,
      "step": 8325
    },
    {
      "epoch": 0.37710224314719665,
      "grad_norm": 0.42449092864990234,
      "learning_rate": 4.4839011993003245e-05,
      "loss": 0.181,
      "num_input_tokens_seen": 13400832,
      "step": 8330
    },
    {
      "epoch": 0.3773285950338396,
      "grad_norm": 0.3408415615558624,
      "learning_rate": 4.4833036613246305e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 13408928,
      "step": 8335
    },
    {
      "epoch": 0.3775549469204826,
      "grad_norm": 0.45135974884033203,
      "learning_rate": 4.482705817498589e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 13416640,
      "step": 8340
    },
    {
      "epoch": 0.3777812988071256,
      "grad_norm": 0.6469824314117432,
      "learning_rate": 4.4821076679143934e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 13425120,
      "step": 8345
    },
    {
      "epoch": 0.3780076506937685,
      "grad_norm": 0.8303060531616211,
      "learning_rate": 4.481509212664288e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 13433088,
      "step": 8350
    },
    {
      "epoch": 0.3782340025804115,
      "grad_norm": 0.3703983426094055,
      "learning_rate": 4.480910451840559e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 13441120,
      "step": 8355
    },
    {
      "epoch": 0.3784603544670545,
      "grad_norm": 0.6296572089195251,
      "learning_rate": 4.480311385535546e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 13448960,
      "step": 8360
    },
    {
      "epoch": 0.37868670635369744,
      "grad_norm": 0.4739671051502228,
      "learning_rate": 4.47971201384163e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 13457312,
      "step": 8365
    },
    {
      "epoch": 0.37891305824034044,
      "grad_norm": 0.6208343505859375,
      "learning_rate": 4.4791123368512446e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 13465792,
      "step": 8370
    },
    {
      "epoch": 0.37913941012698343,
      "grad_norm": 0.35570022463798523,
      "learning_rate": 4.478512354656864e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 13473440,
      "step": 8375
    },
    {
      "epoch": 0.3793657620136264,
      "grad_norm": 0.65306156873703,
      "learning_rate": 4.477912067351016e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 13481600,
      "step": 8380
    },
    {
      "epoch": 0.37959211390026937,
      "grad_norm": 0.8807394504547119,
      "learning_rate": 4.477311475026271e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 13489472,
      "step": 8385
    },
    {
      "epoch": 0.3798184657869123,
      "grad_norm": 0.8990058898925781,
      "learning_rate": 4.476710577775248e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 13497568,
      "step": 8390
    },
    {
      "epoch": 0.3800448176735553,
      "grad_norm": 0.6015975475311279,
      "learning_rate": 4.476109375690612e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 13505952,
      "step": 8395
    },
    {
      "epoch": 0.3802711695601983,
      "grad_norm": 0.8144523501396179,
      "learning_rate": 4.4755078688650784e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 13513696,
      "step": 8400
    },
    {
      "epoch": 0.3802711695601983,
      "eval_loss": 0.1653004288673401,
      "eval_runtime": 405.1237,
      "eval_samples_per_second": 96.936,
      "eval_steps_per_second": 24.235,
      "num_input_tokens_seen": 13513696,
      "step": 8400
    },
    {
      "epoch": 0.38049752144684124,
      "grad_norm": 0.9284787774085999,
      "learning_rate": 4.474906057391406e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 13521664,
      "step": 8405
    },
    {
      "epoch": 0.38072387333348423,
      "grad_norm": 0.7494640350341797,
      "learning_rate": 4.4743039413624e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 13529568,
      "step": 8410
    },
    {
      "epoch": 0.3809502252201272,
      "grad_norm": 0.822478711605072,
      "learning_rate": 4.473701520870916e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 13537472,
      "step": 8415
    },
    {
      "epoch": 0.38117657710677016,
      "grad_norm": 0.6684401035308838,
      "learning_rate": 4.4730987960098544e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 13546080,
      "step": 8420
    },
    {
      "epoch": 0.38140292899341316,
      "grad_norm": 0.42893028259277344,
      "learning_rate": 4.4724957668721635e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 13554272,
      "step": 8425
    },
    {
      "epoch": 0.38162928088005615,
      "grad_norm": 0.6576263904571533,
      "learning_rate": 4.471892433550836e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 13562912,
      "step": 8430
    },
    {
      "epoch": 0.3818556327666991,
      "grad_norm": 0.6143984198570251,
      "learning_rate": 4.471288796138916e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 13571008,
      "step": 8435
    },
    {
      "epoch": 0.3820819846533421,
      "grad_norm": 0.5324165225028992,
      "learning_rate": 4.470684854729491e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 13579232,
      "step": 8440
    },
    {
      "epoch": 0.3823083365399851,
      "grad_norm": 0.42360419034957886,
      "learning_rate": 4.4700806094156955e-05,
      "loss": 0.1832,
      "num_input_tokens_seen": 13586528,
      "step": 8445
    },
    {
      "epoch": 0.382534688426628,
      "grad_norm": 0.5692728757858276,
      "learning_rate": 4.469476060290713e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 13594912,
      "step": 8450
    },
    {
      "epoch": 0.382761040313271,
      "grad_norm": 0.7294952273368835,
      "learning_rate": 4.468871207447772e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 13602368,
      "step": 8455
    },
    {
      "epoch": 0.382987392199914,
      "grad_norm": 0.5304672718048096,
      "learning_rate": 4.4682660509801486e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 13611104,
      "step": 8460
    },
    {
      "epoch": 0.38321374408655695,
      "grad_norm": 0.38856440782546997,
      "learning_rate": 4.467660590981165e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 13618848,
      "step": 8465
    },
    {
      "epoch": 0.38344009597319995,
      "grad_norm": 0.48291775584220886,
      "learning_rate": 4.467054827544191e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 13626784,
      "step": 8470
    },
    {
      "epoch": 0.3836664478598429,
      "grad_norm": 0.415690153837204,
      "learning_rate": 4.4664487607626434e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 13634784,
      "step": 8475
    },
    {
      "epoch": 0.3838927997464859,
      "grad_norm": 0.5362666249275208,
      "learning_rate": 4.4658423907299845e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 13643392,
      "step": 8480
    },
    {
      "epoch": 0.3841191516331289,
      "grad_norm": 0.3715800940990448,
      "learning_rate": 4.465235717539725e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 13651200,
      "step": 8485
    },
    {
      "epoch": 0.3843455035197718,
      "grad_norm": 0.8191201686859131,
      "learning_rate": 4.464628741285421e-05,
      "loss": 0.1682,
      "num_input_tokens_seen": 13659360,
      "step": 8490
    },
    {
      "epoch": 0.3845718554064148,
      "grad_norm": 0.22543519735336304,
      "learning_rate": 4.4640214620606754e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 13668288,
      "step": 8495
    },
    {
      "epoch": 0.3847982072930578,
      "grad_norm": 1.2499003410339355,
      "learning_rate": 4.46341387995914e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 13676320,
      "step": 8500
    },
    {
      "epoch": 0.38502455917970074,
      "grad_norm": 0.4158450663089752,
      "learning_rate": 4.4628059950745106e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 13684224,
      "step": 8505
    },
    {
      "epoch": 0.38525091106634374,
      "grad_norm": 0.6804513931274414,
      "learning_rate": 4.4621978075005297e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 13692320,
      "step": 8510
    },
    {
      "epoch": 0.38547726295298673,
      "grad_norm": 0.5692432522773743,
      "learning_rate": 4.461589317330989e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 13700544,
      "step": 8515
    },
    {
      "epoch": 0.38570361483962967,
      "grad_norm": 1.569525122642517,
      "learning_rate": 4.460980524659724e-05,
      "loss": 0.1824,
      "num_input_tokens_seen": 13708768,
      "step": 8520
    },
    {
      "epoch": 0.38592996672627267,
      "grad_norm": 0.48352211713790894,
      "learning_rate": 4.46037142958062e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 13716704,
      "step": 8525
    },
    {
      "epoch": 0.38615631861291566,
      "grad_norm": 0.7149509787559509,
      "learning_rate": 4.4597620321876046e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 13725024,
      "step": 8530
    },
    {
      "epoch": 0.3863826704995586,
      "grad_norm": 0.5827685594558716,
      "learning_rate": 4.459152332574656e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 13733152,
      "step": 8535
    },
    {
      "epoch": 0.3866090223862016,
      "grad_norm": 0.5303100943565369,
      "learning_rate": 4.4585423308357985e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 13741344,
      "step": 8540
    },
    {
      "epoch": 0.3868353742728446,
      "grad_norm": 0.6840253472328186,
      "learning_rate": 4.457932027065102e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 13749856,
      "step": 8545
    },
    {
      "epoch": 0.38706172615948753,
      "grad_norm": 0.5095581412315369,
      "learning_rate": 4.45732142135668e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 13757504,
      "step": 8550
    },
    {
      "epoch": 0.3872880780461305,
      "grad_norm": 0.46856144070625305,
      "learning_rate": 4.4567105138046986e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 13765952,
      "step": 8555
    },
    {
      "epoch": 0.38751442993277346,
      "grad_norm": 0.859781801700592,
      "learning_rate": 4.456099304503365e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 13775232,
      "step": 8560
    },
    {
      "epoch": 0.38774078181941646,
      "grad_norm": 0.550929605960846,
      "learning_rate": 4.455487793546939e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 13783360,
      "step": 8565
    },
    {
      "epoch": 0.38796713370605945,
      "grad_norm": 0.42531538009643555,
      "learning_rate": 4.454875981029719e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 13791424,
      "step": 8570
    },
    {
      "epoch": 0.3881934855927024,
      "grad_norm": 0.8541380167007446,
      "learning_rate": 4.454263867046057e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 13799296,
      "step": 8575
    },
    {
      "epoch": 0.3884198374793454,
      "grad_norm": 0.6857264041900635,
      "learning_rate": 4.4536514516903484e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 13806976,
      "step": 8580
    },
    {
      "epoch": 0.3886461893659884,
      "grad_norm": 0.33000612258911133,
      "learning_rate": 4.453038735057034e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 13815168,
      "step": 8585
    },
    {
      "epoch": 0.3888725412526313,
      "grad_norm": 0.7575913071632385,
      "learning_rate": 4.4524257172406034e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 13822624,
      "step": 8590
    },
    {
      "epoch": 0.3890988931392743,
      "grad_norm": 1.204803466796875,
      "learning_rate": 4.451812398335592e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 13830720,
      "step": 8595
    },
    {
      "epoch": 0.3893252450259173,
      "grad_norm": 0.5666316151618958,
      "learning_rate": 4.4511987784365805e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 13838880,
      "step": 8600
    },
    {
      "epoch": 0.3893252450259173,
      "eval_loss": 0.16237406432628632,
      "eval_runtime": 404.2927,
      "eval_samples_per_second": 97.135,
      "eval_steps_per_second": 24.284,
      "num_input_tokens_seen": 13838880,
      "step": 8600
    },
    {
      "epoch": 0.38955159691256025,
      "grad_norm": 1.1190747022628784,
      "learning_rate": 4.450584857638197e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 13846720,
      "step": 8605
    },
    {
      "epoch": 0.38977794879920324,
      "grad_norm": 0.7509970664978027,
      "learning_rate": 4.449970636035116e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 13854464,
      "step": 8610
    },
    {
      "epoch": 0.39000430068584624,
      "grad_norm": 0.8191340565681458,
      "learning_rate": 4.4493561137220574e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 13862368,
      "step": 8615
    },
    {
      "epoch": 0.3902306525724892,
      "grad_norm": 0.5628631711006165,
      "learning_rate": 4.44874129079379e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 13870592,
      "step": 8620
    },
    {
      "epoch": 0.3904570044591322,
      "grad_norm": 1.177240252494812,
      "learning_rate": 4.4481261673451255e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 13879712,
      "step": 8625
    },
    {
      "epoch": 0.39068335634577517,
      "grad_norm": 0.8014174103736877,
      "learning_rate": 4.4475107434709245e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 13887552,
      "step": 8630
    },
    {
      "epoch": 0.3909097082324181,
      "grad_norm": 1.3794952630996704,
      "learning_rate": 4.446895019266093e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 13895200,
      "step": 8635
    },
    {
      "epoch": 0.3911360601190611,
      "grad_norm": 0.813446581363678,
      "learning_rate": 4.446278994825583e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 13903904,
      "step": 8640
    },
    {
      "epoch": 0.39136241200570404,
      "grad_norm": 0.5054652690887451,
      "learning_rate": 4.445662670244394e-05,
      "loss": 0.1967,
      "num_input_tokens_seen": 13911936,
      "step": 8645
    },
    {
      "epoch": 0.39158876389234704,
      "grad_norm": 0.4709009528160095,
      "learning_rate": 4.44504604561757e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 13920096,
      "step": 8650
    },
    {
      "epoch": 0.39181511577899003,
      "grad_norm": 0.8598634600639343,
      "learning_rate": 4.4444291210402035e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 13928224,
      "step": 8655
    },
    {
      "epoch": 0.39204146766563297,
      "grad_norm": 1.3132061958312988,
      "learning_rate": 4.443811896607431e-05,
      "loss": 0.2018,
      "num_input_tokens_seen": 13936512,
      "step": 8660
    },
    {
      "epoch": 0.39226781955227596,
      "grad_norm": 0.8615591526031494,
      "learning_rate": 4.443194372414436e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 13944768,
      "step": 8665
    },
    {
      "epoch": 0.39249417143891896,
      "grad_norm": 0.8119844198226929,
      "learning_rate": 4.442576548556449e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 13952768,
      "step": 8670
    },
    {
      "epoch": 0.3927205233255619,
      "grad_norm": 0.2884407341480255,
      "learning_rate": 4.441958425128747e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 13960512,
      "step": 8675
    },
    {
      "epoch": 0.3929468752122049,
      "grad_norm": 0.6480486989021301,
      "learning_rate": 4.4413400022266515e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 13968800,
      "step": 8680
    },
    {
      "epoch": 0.3931732270988479,
      "grad_norm": 0.6197497844696045,
      "learning_rate": 4.4407212799455313e-05,
      "loss": 0.1742,
      "num_input_tokens_seen": 13977056,
      "step": 8685
    },
    {
      "epoch": 0.3933995789854908,
      "grad_norm": 0.6365443468093872,
      "learning_rate": 4.4401022583808003e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 13984832,
      "step": 8690
    },
    {
      "epoch": 0.3936259308721338,
      "grad_norm": 0.45140719413757324,
      "learning_rate": 4.439482937627921e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 13992832,
      "step": 8695
    },
    {
      "epoch": 0.3938522827587768,
      "grad_norm": 0.5307639241218567,
      "learning_rate": 4.4388633177824004e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 14001248,
      "step": 8700
    },
    {
      "epoch": 0.39407863464541976,
      "grad_norm": 0.4731757640838623,
      "learning_rate": 4.4382433989397895e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 14009376,
      "step": 8705
    },
    {
      "epoch": 0.39430498653206275,
      "grad_norm": 0.38961634039878845,
      "learning_rate": 4.4376231811956895e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 14017408,
      "step": 8710
    },
    {
      "epoch": 0.39453133841870575,
      "grad_norm": 0.468350887298584,
      "learning_rate": 4.437002664645745e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 14024960,
      "step": 8715
    },
    {
      "epoch": 0.3947576903053487,
      "grad_norm": 0.46927139163017273,
      "learning_rate": 4.436381849385649e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 14032768,
      "step": 8720
    },
    {
      "epoch": 0.3949840421919917,
      "grad_norm": 0.6432234644889832,
      "learning_rate": 4.435760735511136e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 14041280,
      "step": 8725
    },
    {
      "epoch": 0.3952103940786346,
      "grad_norm": 0.49687066674232483,
      "learning_rate": 4.435139323117992e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 14049408,
      "step": 8730
    },
    {
      "epoch": 0.3954367459652776,
      "grad_norm": 0.6863203048706055,
      "learning_rate": 4.434517612302046e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 14057344,
      "step": 8735
    },
    {
      "epoch": 0.3956630978519206,
      "grad_norm": 0.809170663356781,
      "learning_rate": 4.433895603159174e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 14065216,
      "step": 8740
    },
    {
      "epoch": 0.39588944973856355,
      "grad_norm": 0.5736451745033264,
      "learning_rate": 4.433273295785296e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 14072704,
      "step": 8745
    },
    {
      "epoch": 0.39611580162520654,
      "grad_norm": 0.4910253584384918,
      "learning_rate": 4.432650690276382e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 14080448,
      "step": 8750
    },
    {
      "epoch": 0.39634215351184954,
      "grad_norm": 0.37697482109069824,
      "learning_rate": 4.4320277867284435e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 14088224,
      "step": 8755
    },
    {
      "epoch": 0.3965685053984925,
      "grad_norm": 0.7477352619171143,
      "learning_rate": 4.431404585237541e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 14096128,
      "step": 8760
    },
    {
      "epoch": 0.39679485728513547,
      "grad_norm": 1.0988439321517944,
      "learning_rate": 4.43078108589978e-05,
      "loss": 0.1645,
      "num_input_tokens_seen": 14103904,
      "step": 8765
    },
    {
      "epoch": 0.39702120917177847,
      "grad_norm": 0.40076813101768494,
      "learning_rate": 4.4301572888113116e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 14112384,
      "step": 8770
    },
    {
      "epoch": 0.3972475610584214,
      "grad_norm": 1.4953192472457886,
      "learning_rate": 4.4295331940683337e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 14120320,
      "step": 8775
    },
    {
      "epoch": 0.3974739129450644,
      "grad_norm": 0.5558510422706604,
      "learning_rate": 4.428908801767089e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 14128288,
      "step": 8780
    },
    {
      "epoch": 0.3977002648317074,
      "grad_norm": 1.0820815563201904,
      "learning_rate": 4.428284112003868e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 14136000,
      "step": 8785
    },
    {
      "epoch": 0.39792661671835033,
      "grad_norm": 0.8028689026832581,
      "learning_rate": 4.4276591248750033e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 14144000,
      "step": 8790
    },
    {
      "epoch": 0.39815296860499333,
      "grad_norm": 0.4961450397968292,
      "learning_rate": 4.4270338404768774e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 14151552,
      "step": 8795
    },
    {
      "epoch": 0.3983793204916363,
      "grad_norm": 0.7539381980895996,
      "learning_rate": 4.426408258905917e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 14159072,
      "step": 8800
    },
    {
      "epoch": 0.3983793204916363,
      "eval_loss": 0.16270868480205536,
      "eval_runtime": 403.6513,
      "eval_samples_per_second": 97.289,
      "eval_steps_per_second": 24.323,
      "num_input_tokens_seen": 14159072,
      "step": 8800
    },
    {
      "epoch": 0.39860567237827926,
      "grad_norm": 0.5242312550544739,
      "learning_rate": 4.425782380258594e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 14166592,
      "step": 8805
    },
    {
      "epoch": 0.39883202426492226,
      "grad_norm": 1.3111023902893066,
      "learning_rate": 4.425156204631427e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 14174368,
      "step": 8810
    },
    {
      "epoch": 0.3990583761515652,
      "grad_norm": 0.781936526298523,
      "learning_rate": 4.424529732120981e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 14182528,
      "step": 8815
    },
    {
      "epoch": 0.3992847280382082,
      "grad_norm": 0.5865743160247803,
      "learning_rate": 4.423902962823864e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 14190112,
      "step": 8820
    },
    {
      "epoch": 0.3995110799248512,
      "grad_norm": 0.393027126789093,
      "learning_rate": 4.423275896836733e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 14198080,
      "step": 8825
    },
    {
      "epoch": 0.3997374318114941,
      "grad_norm": 0.7997115254402161,
      "learning_rate": 4.42264853425629e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 14205984,
      "step": 8830
    },
    {
      "epoch": 0.3999637836981371,
      "grad_norm": 0.5499903559684753,
      "learning_rate": 4.4220208751792816e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 14213856,
      "step": 8835
    },
    {
      "epoch": 0.4001901355847801,
      "grad_norm": 0.34689825773239136,
      "learning_rate": 4.421392919702499e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 14221696,
      "step": 8840
    },
    {
      "epoch": 0.40041648747142305,
      "grad_norm": 0.5589877367019653,
      "learning_rate": 4.4207646679227846e-05,
      "loss": 0.1953,
      "num_input_tokens_seen": 14230496,
      "step": 8845
    },
    {
      "epoch": 0.40064283935806605,
      "grad_norm": 0.8579666614532471,
      "learning_rate": 4.42013611993702e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 14238656,
      "step": 8850
    },
    {
      "epoch": 0.40086919124470904,
      "grad_norm": 0.7270665168762207,
      "learning_rate": 4.419507275842135e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 14246496,
      "step": 8855
    },
    {
      "epoch": 0.401095543131352,
      "grad_norm": 0.5680891871452332,
      "learning_rate": 4.418878135735106e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 14254880,
      "step": 8860
    },
    {
      "epoch": 0.401321895017995,
      "grad_norm": 0.3544614613056183,
      "learning_rate": 4.418248699712955e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 14262880,
      "step": 8865
    },
    {
      "epoch": 0.401548246904638,
      "grad_norm": 0.41698911786079407,
      "learning_rate": 4.417618967872748e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 14270944,
      "step": 8870
    },
    {
      "epoch": 0.4017745987912809,
      "grad_norm": 0.35875290632247925,
      "learning_rate": 4.4169889403115985e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 14278592,
      "step": 8875
    },
    {
      "epoch": 0.4020009506779239,
      "grad_norm": 1.1117523908615112,
      "learning_rate": 4.4163586171266627e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 14286624,
      "step": 8880
    },
    {
      "epoch": 0.4022273025645669,
      "grad_norm": 0.41259947419166565,
      "learning_rate": 4.415727998415147e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 14294592,
      "step": 8885
    },
    {
      "epoch": 0.40245365445120984,
      "grad_norm": 0.7346968650817871,
      "learning_rate": 4.4150970842742985e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 14302272,
      "step": 8890
    },
    {
      "epoch": 0.40268000633785284,
      "grad_norm": 0.8840518593788147,
      "learning_rate": 4.4144658748014134e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 14310400,
      "step": 8895
    },
    {
      "epoch": 0.4029063582244958,
      "grad_norm": 0.315913587808609,
      "learning_rate": 4.413834370093831e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 14318560,
      "step": 8900
    },
    {
      "epoch": 0.40313271011113877,
      "grad_norm": 0.7615258693695068,
      "learning_rate": 4.413202570248939e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 14326176,
      "step": 8905
    },
    {
      "epoch": 0.40335906199778176,
      "grad_norm": 0.6964209675788879,
      "learning_rate": 4.412570475364167e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 14334688,
      "step": 8910
    },
    {
      "epoch": 0.4035854138844247,
      "grad_norm": 0.6580433249473572,
      "learning_rate": 4.411938085536994e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 14343744,
      "step": 8915
    },
    {
      "epoch": 0.4038117657710677,
      "grad_norm": 0.4235902428627014,
      "learning_rate": 4.41130540086494e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 14351840,
      "step": 8920
    },
    {
      "epoch": 0.4040381176577107,
      "grad_norm": 0.671129047870636,
      "learning_rate": 4.4106724214455754e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 14359808,
      "step": 8925
    },
    {
      "epoch": 0.40426446954435363,
      "grad_norm": 1.0695825815200806,
      "learning_rate": 4.4100391473765115e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 14367616,
      "step": 8930
    },
    {
      "epoch": 0.4044908214309966,
      "grad_norm": 0.5117464065551758,
      "learning_rate": 4.409405578755408e-05,
      "loss": 0.1858,
      "num_input_tokens_seen": 14375616,
      "step": 8935
    },
    {
      "epoch": 0.4047171733176396,
      "grad_norm": 0.73233562707901,
      "learning_rate": 4.4087717156799705e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 14383840,
      "step": 8940
    },
    {
      "epoch": 0.40494352520428256,
      "grad_norm": 0.48960015177726746,
      "learning_rate": 4.408137558247946e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 14391200,
      "step": 8945
    },
    {
      "epoch": 0.40516987709092556,
      "grad_norm": 1.7434285879135132,
      "learning_rate": 4.4075031065571306e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 14399520,
      "step": 8950
    },
    {
      "epoch": 0.40539622897756855,
      "grad_norm": 1.309962511062622,
      "learning_rate": 4.406868360705366e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 14407840,
      "step": 8955
    },
    {
      "epoch": 0.4056225808642115,
      "grad_norm": 0.5462313294410706,
      "learning_rate": 4.406233320790536e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 14415872,
      "step": 8960
    },
    {
      "epoch": 0.4058489327508545,
      "grad_norm": 0.5488641858100891,
      "learning_rate": 4.4055979869105734e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 14424032,
      "step": 8965
    },
    {
      "epoch": 0.4060752846374975,
      "grad_norm": 0.9647051692008972,
      "learning_rate": 4.404962359163454e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 14432224,
      "step": 8970
    },
    {
      "epoch": 0.4063016365241404,
      "grad_norm": 0.5669388771057129,
      "learning_rate": 4.404326437647199e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 14440000,
      "step": 8975
    },
    {
      "epoch": 0.4065279884107834,
      "grad_norm": 0.4427925944328308,
      "learning_rate": 4.403690222459877e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 14447712,
      "step": 8980
    },
    {
      "epoch": 0.40675434029742635,
      "grad_norm": 0.811549961566925,
      "learning_rate": 4.4030537136995984e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 14456064,
      "step": 8985
    },
    {
      "epoch": 0.40698069218406935,
      "grad_norm": 0.7376483678817749,
      "learning_rate": 4.402416911464523e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 14464896,
      "step": 8990
    },
    {
      "epoch": 0.40720704407071234,
      "grad_norm": 0.6244831681251526,
      "learning_rate": 4.4017798158528516e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 14473120,
      "step": 8995
    },
    {
      "epoch": 0.4074333959573553,
      "grad_norm": 0.6185948252677917,
      "learning_rate": 4.401142426962834e-05,
      "loss": 0.1202,
      "num_input_tokens_seen": 14481184,
      "step": 9000
    },
    {
      "epoch": 0.4074333959573553,
      "eval_loss": 0.16203932464122772,
      "eval_runtime": 404.6748,
      "eval_samples_per_second": 97.043,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 14481184,
      "step": 9000
    },
    {
      "epoch": 0.4076597478439983,
      "grad_norm": 1.1902164220809937,
      "learning_rate": 4.400504744892763e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 14489760,
      "step": 9005
    },
    {
      "epoch": 0.40788609973064127,
      "grad_norm": 0.4726100265979767,
      "learning_rate": 4.399866769740975e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 14497504,
      "step": 9010
    },
    {
      "epoch": 0.4081124516172842,
      "grad_norm": 0.6168621182441711,
      "learning_rate": 4.399228501605859e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 14505600,
      "step": 9015
    },
    {
      "epoch": 0.4083388035039272,
      "grad_norm": 0.7214657068252563,
      "learning_rate": 4.398589940585839e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 14513600,
      "step": 9020
    },
    {
      "epoch": 0.4085651553905702,
      "grad_norm": 0.7070088982582092,
      "learning_rate": 4.3979510867793917e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 14521280,
      "step": 9025
    },
    {
      "epoch": 0.40879150727721314,
      "grad_norm": 0.832161545753479,
      "learning_rate": 4.3973119402850346e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 14528992,
      "step": 9030
    },
    {
      "epoch": 0.40901785916385613,
      "grad_norm": 0.6767604351043701,
      "learning_rate": 4.396672501201334e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 14536960,
      "step": 9035
    },
    {
      "epoch": 0.40924421105049913,
      "grad_norm": 0.36631977558135986,
      "learning_rate": 4.396032769626899e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 14544512,
      "step": 9040
    },
    {
      "epoch": 0.40947056293714207,
      "grad_norm": 0.528286874294281,
      "learning_rate": 4.395392745660384e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 14552320,
      "step": 9045
    },
    {
      "epoch": 0.40969691482378506,
      "grad_norm": 1.2356072664260864,
      "learning_rate": 4.394752429400488e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 14559776,
      "step": 9050
    },
    {
      "epoch": 0.40992326671042806,
      "grad_norm": 1.8562297821044922,
      "learning_rate": 4.394111820945957e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 14568384,
      "step": 9055
    },
    {
      "epoch": 0.410149618597071,
      "grad_norm": 0.4640898108482361,
      "learning_rate": 4.393470920395579e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 14576736,
      "step": 9060
    },
    {
      "epoch": 0.410375970483714,
      "grad_norm": 0.4228704273700714,
      "learning_rate": 4.392829727848192e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 14585152,
      "step": 9065
    },
    {
      "epoch": 0.41060232237035693,
      "grad_norm": 0.6154865622520447,
      "learning_rate": 4.392188243402673e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 14593248,
      "step": 9070
    },
    {
      "epoch": 0.4108286742569999,
      "grad_norm": 1.1559849977493286,
      "learning_rate": 4.391546467157949e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 14601632,
      "step": 9075
    },
    {
      "epoch": 0.4110550261436429,
      "grad_norm": 0.6004852652549744,
      "learning_rate": 4.390904399212988e-05,
      "loss": 0.1842,
      "num_input_tokens_seen": 14609792,
      "step": 9080
    },
    {
      "epoch": 0.41128137803028586,
      "grad_norm": 0.5661202073097229,
      "learning_rate": 4.390262039666807e-05,
      "loss": 0.1857,
      "num_input_tokens_seen": 14617728,
      "step": 9085
    },
    {
      "epoch": 0.41150772991692885,
      "grad_norm": 0.3003944158554077,
      "learning_rate": 4.389619388618464e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 14625248,
      "step": 9090
    },
    {
      "epoch": 0.41173408180357185,
      "grad_norm": 1.1015325784683228,
      "learning_rate": 4.3889764461670655e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 14633184,
      "step": 9095
    },
    {
      "epoch": 0.4119604336902148,
      "grad_norm": 0.3015408515930176,
      "learning_rate": 4.38833321241176e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 14640480,
      "step": 9100
    },
    {
      "epoch": 0.4121867855768578,
      "grad_norm": 0.822347104549408,
      "learning_rate": 4.3876896874517434e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 14648640,
      "step": 9105
    },
    {
      "epoch": 0.4124131374635008,
      "grad_norm": 0.7146496772766113,
      "learning_rate": 4.3870458713862554e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 14656704,
      "step": 9110
    },
    {
      "epoch": 0.4126394893501437,
      "grad_norm": 1.2237889766693115,
      "learning_rate": 4.386401764314579e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 14664576,
      "step": 9115
    },
    {
      "epoch": 0.4128658412367867,
      "grad_norm": 0.3850424587726593,
      "learning_rate": 4.385757366336045e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 14672928,
      "step": 9120
    },
    {
      "epoch": 0.4130921931234297,
      "grad_norm": 0.9423868060112,
      "learning_rate": 4.385112677550027e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 14681408,
      "step": 9125
    },
    {
      "epoch": 0.41331854501007265,
      "grad_norm": 0.7460629343986511,
      "learning_rate": 4.384467698055945e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 14689312,
      "step": 9130
    },
    {
      "epoch": 0.41354489689671564,
      "grad_norm": 0.39459359645843506,
      "learning_rate": 4.383822427953261e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 14697280,
      "step": 9135
    },
    {
      "epoch": 0.41377124878335864,
      "grad_norm": 0.4745575487613678,
      "learning_rate": 4.3831768673414864e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 14705536,
      "step": 9140
    },
    {
      "epoch": 0.4139976006700016,
      "grad_norm": 0.4966196119785309,
      "learning_rate": 4.382531016320173e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 14713088,
      "step": 9145
    },
    {
      "epoch": 0.41422395255664457,
      "grad_norm": 0.5736401677131653,
      "learning_rate": 4.3818848749889184e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 14721280,
      "step": 9150
    },
    {
      "epoch": 0.4144503044432875,
      "grad_norm": 1.009633183479309,
      "learning_rate": 4.381238443447368e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 14730112,
      "step": 9155
    },
    {
      "epoch": 0.4146766563299305,
      "grad_norm": 0.4853321611881256,
      "learning_rate": 4.380591721795208e-05,
      "loss": 0.125,
      "num_input_tokens_seen": 14737984,
      "step": 9160
    },
    {
      "epoch": 0.4149030082165735,
      "grad_norm": 0.3752482831478119,
      "learning_rate": 4.3799447101321723e-05,
      "loss": 0.1223,
      "num_input_tokens_seen": 14746272,
      "step": 9165
    },
    {
      "epoch": 0.41512936010321644,
      "grad_norm": 0.4985067844390869,
      "learning_rate": 4.379297408558036e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 14754560,
      "step": 9170
    },
    {
      "epoch": 0.41535571198985943,
      "grad_norm": 0.6753705739974976,
      "learning_rate": 4.378649817172624e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 14762400,
      "step": 9175
    },
    {
      "epoch": 0.4155820638765024,
      "grad_norm": 0.38949456810951233,
      "learning_rate": 4.378001936075801e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 14770144,
      "step": 9180
    },
    {
      "epoch": 0.41580841576314537,
      "grad_norm": 0.94550621509552,
      "learning_rate": 4.377353765367479e-05,
      "loss": 0.215,
      "num_input_tokens_seen": 14778080,
      "step": 9185
    },
    {
      "epoch": 0.41603476764978836,
      "grad_norm": 0.9649613499641418,
      "learning_rate": 4.376705305147614e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 14786432,
      "step": 9190
    },
    {
      "epoch": 0.41626111953643136,
      "grad_norm": 0.5163437724113464,
      "learning_rate": 4.376056555516206e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 14794624,
      "step": 9195
    },
    {
      "epoch": 0.4164874714230743,
      "grad_norm": 0.8005709648132324,
      "learning_rate": 4.375407516573302e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 14802624,
      "step": 9200
    },
    {
      "epoch": 0.4164874714230743,
      "eval_loss": 0.16176435351371765,
      "eval_runtime": 404.7324,
      "eval_samples_per_second": 97.03,
      "eval_steps_per_second": 24.258,
      "num_input_tokens_seen": 14802624,
      "step": 9200
    },
    {
      "epoch": 0.4167138233097173,
      "grad_norm": 0.3073842227458954,
      "learning_rate": 4.3747581884189913e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 14810656,
      "step": 9205
    },
    {
      "epoch": 0.4169401751963603,
      "grad_norm": 1.484858751296997,
      "learning_rate": 4.374108571153408e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 14819040,
      "step": 9210
    },
    {
      "epoch": 0.4171665270830032,
      "grad_norm": 0.543792724609375,
      "learning_rate": 4.3734586648767316e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 14827616,
      "step": 9215
    },
    {
      "epoch": 0.4173928789696462,
      "grad_norm": 1.0510306358337402,
      "learning_rate": 4.372808469689186e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 14835744,
      "step": 9220
    },
    {
      "epoch": 0.4176192308562892,
      "grad_norm": 1.4351435899734497,
      "learning_rate": 4.372157985691039e-05,
      "loss": 0.2071,
      "num_input_tokens_seen": 14843936,
      "step": 9225
    },
    {
      "epoch": 0.41784558274293215,
      "grad_norm": 0.5774050951004028,
      "learning_rate": 4.371507212982603e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 14852096,
      "step": 9230
    },
    {
      "epoch": 0.41807193462957515,
      "grad_norm": 1.2798511981964111,
      "learning_rate": 4.370856151664236e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 14860096,
      "step": 9235
    },
    {
      "epoch": 0.4182982865162181,
      "grad_norm": 0.5785632133483887,
      "learning_rate": 4.3702048018363404e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 14868352,
      "step": 9240
    },
    {
      "epoch": 0.4185246384028611,
      "grad_norm": 0.49341079592704773,
      "learning_rate": 4.369553163599362e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 14876320,
      "step": 9245
    },
    {
      "epoch": 0.4187509902895041,
      "grad_norm": 0.869645357131958,
      "learning_rate": 4.3689012370537904e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 14885344,
      "step": 9250
    },
    {
      "epoch": 0.418977342176147,
      "grad_norm": 0.2965918481349945,
      "learning_rate": 4.368249022300164e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 14893696,
      "step": 9255
    },
    {
      "epoch": 0.41920369406279,
      "grad_norm": 0.9038086533546448,
      "learning_rate": 4.367596519439059e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 14901472,
      "step": 9260
    },
    {
      "epoch": 0.419430045949433,
      "grad_norm": 0.37798380851745605,
      "learning_rate": 4.366943728571101e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 14909888,
      "step": 9265
    },
    {
      "epoch": 0.41965639783607595,
      "grad_norm": 1.1248879432678223,
      "learning_rate": 4.366290649796959e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 14917920,
      "step": 9270
    },
    {
      "epoch": 0.41988274972271894,
      "grad_norm": 0.5755655765533447,
      "learning_rate": 4.3656372832173456e-05,
      "loss": 0.1782,
      "num_input_tokens_seen": 14926144,
      "step": 9275
    },
    {
      "epoch": 0.42010910160936193,
      "grad_norm": 1.4367039203643799,
      "learning_rate": 4.364983628933017e-05,
      "loss": 0.2035,
      "num_input_tokens_seen": 14934976,
      "step": 9280
    },
    {
      "epoch": 0.4203354534960049,
      "grad_norm": 0.5560685992240906,
      "learning_rate": 4.364329687044777e-05,
      "loss": 0.1871,
      "num_input_tokens_seen": 14943072,
      "step": 9285
    },
    {
      "epoch": 0.42056180538264787,
      "grad_norm": 0.3961128294467926,
      "learning_rate": 4.36367545765347e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 14951392,
      "step": 9290
    },
    {
      "epoch": 0.42078815726929086,
      "grad_norm": 0.3098580539226532,
      "learning_rate": 4.363020940859988e-05,
      "loss": 0.1982,
      "num_input_tokens_seen": 14959296,
      "step": 9295
    },
    {
      "epoch": 0.4210145091559338,
      "grad_norm": 1.7903553247451782,
      "learning_rate": 4.362366136765263e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 14966848,
      "step": 9300
    },
    {
      "epoch": 0.4212408610425768,
      "grad_norm": 0.6089984178543091,
      "learning_rate": 4.361711045470278e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 14974880,
      "step": 9305
    },
    {
      "epoch": 0.4214672129292198,
      "grad_norm": 0.9697222709655762,
      "learning_rate": 4.3610556670760524e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 14982560,
      "step": 9310
    },
    {
      "epoch": 0.42169356481586273,
      "grad_norm": 0.5366050004959106,
      "learning_rate": 4.360400001683657e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 14990336,
      "step": 9315
    },
    {
      "epoch": 0.4219199167025057,
      "grad_norm": 0.8138255476951599,
      "learning_rate": 4.3597440493942e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 14998560,
      "step": 9320
    },
    {
      "epoch": 0.42214626858914867,
      "grad_norm": 0.7999310493469238,
      "learning_rate": 4.3590878103088405e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 15006464,
      "step": 9325
    },
    {
      "epoch": 0.42237262047579166,
      "grad_norm": 0.4301348030567169,
      "learning_rate": 4.358431284528779e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 15014624,
      "step": 9330
    },
    {
      "epoch": 0.42259897236243466,
      "grad_norm": 0.905788779258728,
      "learning_rate": 4.357774472155257e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 15022880,
      "step": 9335
    },
    {
      "epoch": 0.4228253242490776,
      "grad_norm": 1.0231939554214478,
      "learning_rate": 4.3571173732895664e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 15030880,
      "step": 9340
    },
    {
      "epoch": 0.4230516761357206,
      "grad_norm": 1.526977300643921,
      "learning_rate": 4.356459988033039e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 15038848,
      "step": 9345
    },
    {
      "epoch": 0.4232780280223636,
      "grad_norm": 0.4573092758655548,
      "learning_rate": 4.355802316487051e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 15047168,
      "step": 9350
    },
    {
      "epoch": 0.4235043799090065,
      "grad_norm": 0.8344992995262146,
      "learning_rate": 4.355144358753025e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 15055648,
      "step": 9355
    },
    {
      "epoch": 0.4237307317956495,
      "grad_norm": 1.067468523979187,
      "learning_rate": 4.354486114932425e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 15063872,
      "step": 9360
    },
    {
      "epoch": 0.4239570836822925,
      "grad_norm": 0.5343531370162964,
      "learning_rate": 4.353827585126762e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 15072704,
      "step": 9365
    },
    {
      "epoch": 0.42418343556893545,
      "grad_norm": 0.4687281847000122,
      "learning_rate": 4.353168769437588e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 15081216,
      "step": 9370
    },
    {
      "epoch": 0.42440978745557845,
      "grad_norm": 0.8251868486404419,
      "learning_rate": 4.3525096679665014e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 15089120,
      "step": 9375
    },
    {
      "epoch": 0.42463613934222144,
      "grad_norm": 0.4119417369365692,
      "learning_rate": 4.351850280815144e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 15097216,
      "step": 9380
    },
    {
      "epoch": 0.4248624912288644,
      "grad_norm": 0.49264875054359436,
      "learning_rate": 4.3511906080852014e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 15105312,
      "step": 9385
    },
    {
      "epoch": 0.4250888431155074,
      "grad_norm": 0.29327404499053955,
      "learning_rate": 4.350530649878404e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 15112608,
      "step": 9390
    },
    {
      "epoch": 0.42531519500215037,
      "grad_norm": 0.44510260224342346,
      "learning_rate": 4.3498704062965246e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 15121088,
      "step": 9395
    },
    {
      "epoch": 0.4255415468887933,
      "grad_norm": 1.0151599645614624,
      "learning_rate": 4.3492098774413815e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 15129216,
      "step": 9400
    },
    {
      "epoch": 0.4255415468887933,
      "eval_loss": 0.16082808375358582,
      "eval_runtime": 404.3356,
      "eval_samples_per_second": 97.125,
      "eval_steps_per_second": 24.282,
      "num_input_tokens_seen": 15129216,
      "step": 9400
    },
    {
      "epoch": 0.4257678987754363,
      "grad_norm": 0.964314877986908,
      "learning_rate": 4.3485490634148375e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 15137024,
      "step": 9405
    },
    {
      "epoch": 0.42599425066207924,
      "grad_norm": 0.30300483107566833,
      "learning_rate": 4.347887964318797e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 15145120,
      "step": 9410
    },
    {
      "epoch": 0.42622060254872224,
      "grad_norm": 0.7366275191307068,
      "learning_rate": 4.34722658025521e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 15152800,
      "step": 9415
    },
    {
      "epoch": 0.42644695443536523,
      "grad_norm": 0.8115053772926331,
      "learning_rate": 4.346564911326071e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 15160768,
      "step": 9420
    },
    {
      "epoch": 0.4266733063220082,
      "grad_norm": 0.6620774269104004,
      "learning_rate": 4.345902957633418e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 15168448,
      "step": 9425
    },
    {
      "epoch": 0.42689965820865117,
      "grad_norm": 0.9661250710487366,
      "learning_rate": 4.345240719279331e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 15175744,
      "step": 9430
    },
    {
      "epoch": 0.42712601009529416,
      "grad_norm": 0.47273996472358704,
      "learning_rate": 4.3445781963659374e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 15183200,
      "step": 9435
    },
    {
      "epoch": 0.4273523619819371,
      "grad_norm": 0.5908329486846924,
      "learning_rate": 4.3439153889954045e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 15190848,
      "step": 9440
    },
    {
      "epoch": 0.4275787138685801,
      "grad_norm": 0.34945589303970337,
      "learning_rate": 4.343252297269946e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 15199136,
      "step": 9445
    },
    {
      "epoch": 0.4278050657552231,
      "grad_norm": 1.459986925125122,
      "learning_rate": 4.342588921291821e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 15206720,
      "step": 9450
    },
    {
      "epoch": 0.42803141764186603,
      "grad_norm": 0.6412979960441589,
      "learning_rate": 4.341925261163328e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 15214464,
      "step": 9455
    },
    {
      "epoch": 0.428257769528509,
      "grad_norm": 0.266576886177063,
      "learning_rate": 4.341261316986813e-05,
      "loss": 0.1091,
      "num_input_tokens_seen": 15222624,
      "step": 9460
    },
    {
      "epoch": 0.428484121415152,
      "grad_norm": 0.7995045185089111,
      "learning_rate": 4.340597088864664e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 15230880,
      "step": 9465
    },
    {
      "epoch": 0.42871047330179496,
      "grad_norm": 0.4376918077468872,
      "learning_rate": 4.339932576899313e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 15238816,
      "step": 9470
    },
    {
      "epoch": 0.42893682518843795,
      "grad_norm": 0.5411761999130249,
      "learning_rate": 4.3392677811932375e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 15246848,
      "step": 9475
    },
    {
      "epoch": 0.42916317707508095,
      "grad_norm": 0.4562828838825226,
      "learning_rate": 4.338602701848956e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 15255168,
      "step": 9480
    },
    {
      "epoch": 0.4293895289617239,
      "grad_norm": 0.4109184443950653,
      "learning_rate": 4.337937338969033e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 15262976,
      "step": 9485
    },
    {
      "epoch": 0.4296158808483669,
      "grad_norm": 0.3667792081832886,
      "learning_rate": 4.337271692656075e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 15270784,
      "step": 9490
    },
    {
      "epoch": 0.4298422327350098,
      "grad_norm": 0.35417139530181885,
      "learning_rate": 4.336605763012733e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 15278496,
      "step": 9495
    },
    {
      "epoch": 0.4300685846216528,
      "grad_norm": 1.151497721672058,
      "learning_rate": 4.3359395501417026e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 15286848,
      "step": 9500
    },
    {
      "epoch": 0.4302949365082958,
      "grad_norm": 0.40063223242759705,
      "learning_rate": 4.335273054145722e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 15294496,
      "step": 9505
    },
    {
      "epoch": 0.43052128839493875,
      "grad_norm": 0.28129562735557556,
      "learning_rate": 4.334606275127572e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 15302048,
      "step": 9510
    },
    {
      "epoch": 0.43074764028158175,
      "grad_norm": 0.3865644037723541,
      "learning_rate": 4.33393921319008e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 15309664,
      "step": 9515
    },
    {
      "epoch": 0.43097399216822474,
      "grad_norm": 1.1998980045318604,
      "learning_rate": 4.3332718684361146e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 15317248,
      "step": 9520
    },
    {
      "epoch": 0.4312003440548677,
      "grad_norm": 0.6579251289367676,
      "learning_rate": 4.332604240968588e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 15325344,
      "step": 9525
    },
    {
      "epoch": 0.4314266959415107,
      "grad_norm": 0.7439170479774475,
      "learning_rate": 4.331936330890459e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 15333888,
      "step": 9530
    },
    {
      "epoch": 0.43165304782815367,
      "grad_norm": 0.39887475967407227,
      "learning_rate": 4.331268138304725e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 15342816,
      "step": 9535
    },
    {
      "epoch": 0.4318793997147966,
      "grad_norm": 0.4112294912338257,
      "learning_rate": 4.330599663314431e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 15350784,
      "step": 9540
    },
    {
      "epoch": 0.4321057516014396,
      "grad_norm": 0.674948513507843,
      "learning_rate": 4.329930906022665e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 15358688,
      "step": 9545
    },
    {
      "epoch": 0.4323321034880826,
      "grad_norm": 0.3603665232658386,
      "learning_rate": 4.3292618665325564e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 15366016,
      "step": 9550
    },
    {
      "epoch": 0.43255845537472554,
      "grad_norm": 0.552838146686554,
      "learning_rate": 4.3285925449472796e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 15373696,
      "step": 9555
    },
    {
      "epoch": 0.43278480726136853,
      "grad_norm": 1.1772593259811401,
      "learning_rate": 4.327922941370054e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 15381792,
      "step": 9560
    },
    {
      "epoch": 0.4330111591480115,
      "grad_norm": 0.9719095230102539,
      "learning_rate": 4.3272530559041384e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 15389536,
      "step": 9565
    },
    {
      "epoch": 0.43323751103465447,
      "grad_norm": 0.8534179329872131,
      "learning_rate": 4.32658288865284e-05,
      "loss": 0.1821,
      "num_input_tokens_seen": 15397312,
      "step": 9570
    },
    {
      "epoch": 0.43346386292129746,
      "grad_norm": 0.5232947468757629,
      "learning_rate": 4.325912439719505e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 15405248,
      "step": 9575
    },
    {
      "epoch": 0.4336902148079404,
      "grad_norm": 0.9207463264465332,
      "learning_rate": 4.3252417092075266e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 15412992,
      "step": 9580
    },
    {
      "epoch": 0.4339165666945834,
      "grad_norm": 0.8536246418952942,
      "learning_rate": 4.3245706972203385e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 15420704,
      "step": 9585
    },
    {
      "epoch": 0.4341429185812264,
      "grad_norm": 1.5283271074295044,
      "learning_rate": 4.323899403861421e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 15428896,
      "step": 9590
    },
    {
      "epoch": 0.43436927046786933,
      "grad_norm": 0.5630621314048767,
      "learning_rate": 4.3232278292342935e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 15436736,
      "step": 9595
    },
    {
      "epoch": 0.4345956223545123,
      "grad_norm": 0.4780498743057251,
      "learning_rate": 4.322555973442524e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 15445632,
      "step": 9600
    },
    {
      "epoch": 0.4345956223545123,
      "eval_loss": 0.15997463464736938,
      "eval_runtime": 404.7618,
      "eval_samples_per_second": 97.022,
      "eval_steps_per_second": 24.256,
      "num_input_tokens_seen": 15445632,
      "step": 9600
    },
    {
      "epoch": 0.4348219742411553,
      "grad_norm": 0.4194534420967102,
      "learning_rate": 4.3218838365897184e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 15453632,
      "step": 9605
    },
    {
      "epoch": 0.43504832612779826,
      "grad_norm": 0.7919234037399292,
      "learning_rate": 4.3212114187795306e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 15461248,
      "step": 9610
    },
    {
      "epoch": 0.43527467801444125,
      "grad_norm": 0.9898831248283386,
      "learning_rate": 4.320538720115656e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 15469408,
      "step": 9615
    },
    {
      "epoch": 0.43550102990108425,
      "grad_norm": 0.8512096405029297,
      "learning_rate": 4.319865740701831e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 15477120,
      "step": 9620
    },
    {
      "epoch": 0.4357273817877272,
      "grad_norm": 0.5510858297348022,
      "learning_rate": 4.3191924806418396e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 15485056,
      "step": 9625
    },
    {
      "epoch": 0.4359537336743702,
      "grad_norm": 0.3130579888820648,
      "learning_rate": 4.318518940039507e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 15493280,
      "step": 9630
    },
    {
      "epoch": 0.4361800855610132,
      "grad_norm": 0.958832859992981,
      "learning_rate": 4.3178451189987e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 15501216,
      "step": 9635
    },
    {
      "epoch": 0.4364064374476561,
      "grad_norm": 0.7366960048675537,
      "learning_rate": 4.3171710176233315e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 15509088,
      "step": 9640
    },
    {
      "epoch": 0.4366327893342991,
      "grad_norm": 0.6151371002197266,
      "learning_rate": 4.316496636017355e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 15516992,
      "step": 9645
    },
    {
      "epoch": 0.43685914122094205,
      "grad_norm": 0.5041120052337646,
      "learning_rate": 4.315821974284771e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 15525024,
      "step": 9650
    },
    {
      "epoch": 0.43708549310758504,
      "grad_norm": 0.4837587773799896,
      "learning_rate": 4.315147032529619e-05,
      "loss": 0.1259,
      "num_input_tokens_seen": 15532992,
      "step": 9655
    },
    {
      "epoch": 0.43731184499422804,
      "grad_norm": 0.6327344179153442,
      "learning_rate": 4.3144718108559845e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 15540832,
      "step": 9660
    },
    {
      "epoch": 0.437538196880871,
      "grad_norm": 0.4178714454174042,
      "learning_rate": 4.3137963093679945e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 15548704,
      "step": 9665
    },
    {
      "epoch": 0.437764548767514,
      "grad_norm": 0.6620957851409912,
      "learning_rate": 4.31312052816982e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 15556960,
      "step": 9670
    },
    {
      "epoch": 0.43799090065415697,
      "grad_norm": 0.34369903802871704,
      "learning_rate": 4.312444467365675e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 15564896,
      "step": 9675
    },
    {
      "epoch": 0.4382172525407999,
      "grad_norm": 1.3488346338272095,
      "learning_rate": 4.311768127059816e-05,
      "loss": 0.1118,
      "num_input_tokens_seen": 15573856,
      "step": 9680
    },
    {
      "epoch": 0.4384436044274429,
      "grad_norm": 1.131425142288208,
      "learning_rate": 4.3110915073565444e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 15582016,
      "step": 9685
    },
    {
      "epoch": 0.4386699563140859,
      "grad_norm": 0.276309609413147,
      "learning_rate": 4.310414608360203e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 15590048,
      "step": 9690
    },
    {
      "epoch": 0.43889630820072884,
      "grad_norm": 0.5595974922180176,
      "learning_rate": 4.309737430175177e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 15598560,
      "step": 9695
    },
    {
      "epoch": 0.43912266008737183,
      "grad_norm": 0.3402378559112549,
      "learning_rate": 4.309059972905897e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 15606432,
      "step": 9700
    },
    {
      "epoch": 0.4393490119740148,
      "grad_norm": 0.3294239342212677,
      "learning_rate": 4.308382236656836e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 15614496,
      "step": 9705
    },
    {
      "epoch": 0.43957536386065776,
      "grad_norm": 0.40918654203414917,
      "learning_rate": 4.307704221532507e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 15622368,
      "step": 9710
    },
    {
      "epoch": 0.43980171574730076,
      "grad_norm": 0.6289231181144714,
      "learning_rate": 4.307025927637471e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 15630112,
      "step": 9715
    },
    {
      "epoch": 0.44002806763394375,
      "grad_norm": 0.6079529523849487,
      "learning_rate": 4.306347355076328e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 15638048,
      "step": 9720
    },
    {
      "epoch": 0.4402544195205867,
      "grad_norm": 0.737133264541626,
      "learning_rate": 4.305668503953724e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 15646080,
      "step": 9725
    },
    {
      "epoch": 0.4404807714072297,
      "grad_norm": 0.7138616442680359,
      "learning_rate": 4.3049893743743436e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 15654240,
      "step": 9730
    },
    {
      "epoch": 0.4407071232938726,
      "grad_norm": 0.8070418834686279,
      "learning_rate": 4.304309966442919e-05,
      "loss": 0.1865,
      "num_input_tokens_seen": 15662240,
      "step": 9735
    },
    {
      "epoch": 0.4409334751805156,
      "grad_norm": 0.588803768157959,
      "learning_rate": 4.303630280264224e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 15670240,
      "step": 9740
    },
    {
      "epoch": 0.4411598270671586,
      "grad_norm": 0.37864771485328674,
      "learning_rate": 4.302950315943074e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 15678016,
      "step": 9745
    },
    {
      "epoch": 0.44138617895380156,
      "grad_norm": 0.9664386510848999,
      "learning_rate": 4.3022700735843275e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 15685824,
      "step": 9750
    },
    {
      "epoch": 0.44161253084044455,
      "grad_norm": 0.488829642534256,
      "learning_rate": 4.301589553292887e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 15693824,
      "step": 9755
    },
    {
      "epoch": 0.44183888272708755,
      "grad_norm": 1.08351469039917,
      "learning_rate": 4.300908755173697e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 15702080,
      "step": 9760
    },
    {
      "epoch": 0.4420652346137305,
      "grad_norm": 0.8030334115028381,
      "learning_rate": 4.300227679331745e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 15711680,
      "step": 9765
    },
    {
      "epoch": 0.4422915865003735,
      "grad_norm": 0.927640974521637,
      "learning_rate": 4.299546325872063e-05,
      "loss": 0.2023,
      "num_input_tokens_seen": 15719968,
      "step": 9770
    },
    {
      "epoch": 0.4425179383870165,
      "grad_norm": 0.7905915379524231,
      "learning_rate": 4.2988646948997225e-05,
      "loss": 0.2019,
      "num_input_tokens_seen": 15728160,
      "step": 9775
    },
    {
      "epoch": 0.4427442902736594,
      "grad_norm": 1.0886270999908447,
      "learning_rate": 4.29818278651984e-05,
      "loss": 0.1838,
      "num_input_tokens_seen": 15735968,
      "step": 9780
    },
    {
      "epoch": 0.4429706421603024,
      "grad_norm": 0.34204012155532837,
      "learning_rate": 4.297500600837574e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 15744224,
      "step": 9785
    },
    {
      "epoch": 0.4431969940469454,
      "grad_norm": 0.7974473237991333,
      "learning_rate": 4.2968181379581276e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 15752000,
      "step": 9790
    },
    {
      "epoch": 0.44342334593358834,
      "grad_norm": 0.7679380774497986,
      "learning_rate": 4.296135397986743e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 15760032,
      "step": 9795
    },
    {
      "epoch": 0.44364969782023134,
      "grad_norm": 0.6230566501617432,
      "learning_rate": 4.295452381028709e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 15768224,
      "step": 9800
    },
    {
      "epoch": 0.44364969782023134,
      "eval_loss": 0.16087175905704498,
      "eval_runtime": 404.6069,
      "eval_samples_per_second": 97.06,
      "eval_steps_per_second": 24.266,
      "num_input_tokens_seen": 15768224,
      "step": 9800
    },
    {
      "epoch": 0.44387604970687433,
      "grad_norm": 0.41424524784088135,
      "learning_rate": 4.294769087189354e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 15777024,
      "step": 9805
    },
    {
      "epoch": 0.44410240159351727,
      "grad_norm": 0.9212641716003418,
      "learning_rate": 4.294085516574052e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 15785216,
      "step": 9810
    },
    {
      "epoch": 0.44432875348016027,
      "grad_norm": 1.1368461847305298,
      "learning_rate": 4.2934016692882176e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 15793152,
      "step": 9815
    },
    {
      "epoch": 0.4445551053668032,
      "grad_norm": 0.6289729475975037,
      "learning_rate": 4.292717545437308e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 15800704,
      "step": 9820
    },
    {
      "epoch": 0.4447814572534462,
      "grad_norm": 0.5414636731147766,
      "learning_rate": 4.292033145126825e-05,
      "loss": 0.1815,
      "num_input_tokens_seen": 15808448,
      "step": 9825
    },
    {
      "epoch": 0.4450078091400892,
      "grad_norm": 0.5186102390289307,
      "learning_rate": 4.29134846846231e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 15816480,
      "step": 9830
    },
    {
      "epoch": 0.44523416102673213,
      "grad_norm": 0.3532852530479431,
      "learning_rate": 4.29066351554935e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 15824160,
      "step": 9835
    },
    {
      "epoch": 0.44546051291337513,
      "grad_norm": 0.7102212905883789,
      "learning_rate": 4.289978286493574e-05,
      "loss": 0.2053,
      "num_input_tokens_seen": 15831968,
      "step": 9840
    },
    {
      "epoch": 0.4456868648000181,
      "grad_norm": 0.699080228805542,
      "learning_rate": 4.28929278140065e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 15839648,
      "step": 9845
    },
    {
      "epoch": 0.44591321668666106,
      "grad_norm": 0.43202680349349976,
      "learning_rate": 4.288607000376295e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 15847648,
      "step": 9850
    },
    {
      "epoch": 0.44613956857330406,
      "grad_norm": 0.42682021856307983,
      "learning_rate": 4.2879209435262624e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 15855552,
      "step": 9855
    },
    {
      "epoch": 0.44636592045994705,
      "grad_norm": 0.5047386884689331,
      "learning_rate": 4.287234610956353e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 15863744,
      "step": 9860
    },
    {
      "epoch": 0.44659227234659,
      "grad_norm": 0.5410216450691223,
      "learning_rate": 4.2865480027724056e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 15871840,
      "step": 9865
    },
    {
      "epoch": 0.446818624233233,
      "grad_norm": 0.5238369703292847,
      "learning_rate": 4.285861119080306e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 15879520,
      "step": 9870
    },
    {
      "epoch": 0.447044976119876,
      "grad_norm": 0.7129098773002625,
      "learning_rate": 4.2851739599859784e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 15887264,
      "step": 9875
    },
    {
      "epoch": 0.4472713280065189,
      "grad_norm": 0.5682781934738159,
      "learning_rate": 4.2844865255953934e-05,
      "loss": 0.2022,
      "num_input_tokens_seen": 15895552,
      "step": 9880
    },
    {
      "epoch": 0.4474976798931619,
      "grad_norm": 0.8112343549728394,
      "learning_rate": 4.2837988160145605e-05,
      "loss": 0.1875,
      "num_input_tokens_seen": 15903392,
      "step": 9885
    },
    {
      "epoch": 0.4477240317798049,
      "grad_norm": 0.9711675047874451,
      "learning_rate": 4.2831108313495336e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 15911200,
      "step": 9890
    },
    {
      "epoch": 0.44795038366644785,
      "grad_norm": 0.3652757704257965,
      "learning_rate": 4.282422571706408e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 15919040,
      "step": 9895
    },
    {
      "epoch": 0.44817673555309084,
      "grad_norm": 0.33025413751602173,
      "learning_rate": 4.281734037191323e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 15926784,
      "step": 9900
    },
    {
      "epoch": 0.4484030874397338,
      "grad_norm": 0.45673444867134094,
      "learning_rate": 4.281045227910459e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 15935424,
      "step": 9905
    },
    {
      "epoch": 0.4486294393263768,
      "grad_norm": 0.8156107068061829,
      "learning_rate": 4.280356143970038e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 15943648,
      "step": 9910
    },
    {
      "epoch": 0.4488557912130198,
      "grad_norm": 0.3310624361038208,
      "learning_rate": 4.279666785476327e-05,
      "loss": 0.1006,
      "num_input_tokens_seen": 15951808,
      "step": 9915
    },
    {
      "epoch": 0.4490821430996627,
      "grad_norm": 0.726715087890625,
      "learning_rate": 4.2789771525356325e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 15960640,
      "step": 9920
    },
    {
      "epoch": 0.4493084949863057,
      "grad_norm": 0.9126155376434326,
      "learning_rate": 4.2782872452543056e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 15968320,
      "step": 9925
    },
    {
      "epoch": 0.4495348468729487,
      "grad_norm": 0.6708203554153442,
      "learning_rate": 4.2775970637387376e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 15975968,
      "step": 9930
    },
    {
      "epoch": 0.44976119875959164,
      "grad_norm": 0.5741444230079651,
      "learning_rate": 4.276906608095363e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 15983776,
      "step": 9935
    },
    {
      "epoch": 0.44998755064623464,
      "grad_norm": 0.4691399931907654,
      "learning_rate": 4.276215878430661e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 15991360,
      "step": 9940
    },
    {
      "epoch": 0.45021390253287763,
      "grad_norm": 0.6675894856452942,
      "learning_rate": 4.275524874851149e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 15999872,
      "step": 9945
    },
    {
      "epoch": 0.45044025441952057,
      "grad_norm": 0.7842504382133484,
      "learning_rate": 4.274833597463388e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 16007808,
      "step": 9950
    },
    {
      "epoch": 0.45066660630616356,
      "grad_norm": 0.6619510650634766,
      "learning_rate": 4.2741420463739824e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 16016064,
      "step": 9955
    },
    {
      "epoch": 0.45089295819280656,
      "grad_norm": 1.2105149030685425,
      "learning_rate": 4.273450221689578e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 16024032,
      "step": 9960
    },
    {
      "epoch": 0.4511193100794495,
      "grad_norm": 0.7704189419746399,
      "learning_rate": 4.272758123516863e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 16031840,
      "step": 9965
    },
    {
      "epoch": 0.4513456619660925,
      "grad_norm": 0.6919211745262146,
      "learning_rate": 4.272065751962567e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 16039680,
      "step": 9970
    },
    {
      "epoch": 0.4515720138527355,
      "grad_norm": 1.721585988998413,
      "learning_rate": 4.271373107133464e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 16049248,
      "step": 9975
    },
    {
      "epoch": 0.4517983657393784,
      "grad_norm": 0.8893747329711914,
      "learning_rate": 4.270680189136366e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 16057216,
      "step": 9980
    },
    {
      "epoch": 0.4520247176260214,
      "grad_norm": 0.8310621380805969,
      "learning_rate": 4.269986998078132e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 16065824,
      "step": 9985
    },
    {
      "epoch": 0.45225106951266436,
      "grad_norm": 0.6556004285812378,
      "learning_rate": 4.2692935340656595e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 16073792,
      "step": 9990
    },
    {
      "epoch": 0.45247742139930736,
      "grad_norm": 0.5099279284477234,
      "learning_rate": 4.26859979720589e-05,
      "loss": 0.1904,
      "num_input_tokens_seen": 16081952,
      "step": 9995
    },
    {
      "epoch": 0.45270377328595035,
      "grad_norm": 0.6521613001823425,
      "learning_rate": 4.267905787605806e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 16090112,
      "step": 10000
    },
    {
      "epoch": 0.45270377328595035,
      "eval_loss": 0.15979456901550293,
      "eval_runtime": 405.3729,
      "eval_samples_per_second": 96.876,
      "eval_steps_per_second": 24.22,
      "num_input_tokens_seen": 16090112,
      "step": 10000
    },
    {
      "epoch": 0.4529301251725933,
      "grad_norm": 0.47494420409202576,
      "learning_rate": 4.267211505372433e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 16098208,
      "step": 10005
    },
    {
      "epoch": 0.4531564770592363,
      "grad_norm": 0.7702585458755493,
      "learning_rate": 4.266516950612837e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 16105536,
      "step": 10010
    },
    {
      "epoch": 0.4533828289458793,
      "grad_norm": 0.6061206459999084,
      "learning_rate": 4.265822123434128e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 16113152,
      "step": 10015
    },
    {
      "epoch": 0.4536091808325222,
      "grad_norm": 0.48283424973487854,
      "learning_rate": 4.265127023943457e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 16121088,
      "step": 10020
    },
    {
      "epoch": 0.4538355327191652,
      "grad_norm": 0.7765209078788757,
      "learning_rate": 4.2644316522480176e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 16129376,
      "step": 10025
    },
    {
      "epoch": 0.4540618846058082,
      "grad_norm": 1.0382397174835205,
      "learning_rate": 4.263736008455044e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 16137600,
      "step": 10030
    },
    {
      "epoch": 0.45428823649245115,
      "grad_norm": 0.6164340972900391,
      "learning_rate": 4.2630400926718125e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 16146176,
      "step": 10035
    },
    {
      "epoch": 0.45451458837909414,
      "grad_norm": 0.3876033127307892,
      "learning_rate": 4.262343905005644e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 16154240,
      "step": 10040
    },
    {
      "epoch": 0.45474094026573714,
      "grad_norm": 0.3401319086551666,
      "learning_rate": 4.261647445563897e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 16161824,
      "step": 10045
    },
    {
      "epoch": 0.4549672921523801,
      "grad_norm": 1.0863572359085083,
      "learning_rate": 4.260950714453976e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 16169952,
      "step": 10050
    },
    {
      "epoch": 0.45519364403902307,
      "grad_norm": 0.39421606063842773,
      "learning_rate": 4.2602537117833266e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 16178048,
      "step": 10055
    },
    {
      "epoch": 0.45541999592566607,
      "grad_norm": 0.6143089532852173,
      "learning_rate": 4.259556437659433e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 16185984,
      "step": 10060
    },
    {
      "epoch": 0.455646347812309,
      "grad_norm": 0.5204653143882751,
      "learning_rate": 4.258858892189825e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 16193312,
      "step": 10065
    },
    {
      "epoch": 0.455872699698952,
      "grad_norm": 1.357056975364685,
      "learning_rate": 4.2581610754820725e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 16201632,
      "step": 10070
    },
    {
      "epoch": 0.45609905158559494,
      "grad_norm": 0.7178350687026978,
      "learning_rate": 4.2574629876437876e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 16210144,
      "step": 10075
    },
    {
      "epoch": 0.45632540347223793,
      "grad_norm": 0.7980347871780396,
      "learning_rate": 4.256764628782625e-05,
      "loss": 0.1799,
      "num_input_tokens_seen": 16218336,
      "step": 10080
    },
    {
      "epoch": 0.45655175535888093,
      "grad_norm": 0.9693030714988708,
      "learning_rate": 4.256065999006279e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 16225856,
      "step": 10085
    },
    {
      "epoch": 0.45677810724552387,
      "grad_norm": 0.7622657418251038,
      "learning_rate": 4.2553670984224885e-05,
      "loss": 0.1974,
      "num_input_tokens_seen": 16233952,
      "step": 10090
    },
    {
      "epoch": 0.45700445913216686,
      "grad_norm": 0.47521448135375977,
      "learning_rate": 4.254667927139032e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 16241792,
      "step": 10095
    },
    {
      "epoch": 0.45723081101880986,
      "grad_norm": 1.0154194831848145,
      "learning_rate": 4.2539684852637295e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 16250272,
      "step": 10100
    },
    {
      "epoch": 0.4574571629054528,
      "grad_norm": 0.8741740584373474,
      "learning_rate": 4.253268772904446e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 16258368,
      "step": 10105
    },
    {
      "epoch": 0.4576835147920958,
      "grad_norm": 0.39365914463996887,
      "learning_rate": 4.252568790169085e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 16266240,
      "step": 10110
    },
    {
      "epoch": 0.4579098666787388,
      "grad_norm": 0.8143216371536255,
      "learning_rate": 4.251868537165592e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 16274400,
      "step": 10115
    },
    {
      "epoch": 0.4581362185653817,
      "grad_norm": 1.04533052444458,
      "learning_rate": 4.251168014001955e-05,
      "loss": 0.1922,
      "num_input_tokens_seen": 16282464,
      "step": 10120
    },
    {
      "epoch": 0.4583625704520247,
      "grad_norm": 0.6800335645675659,
      "learning_rate": 4.250467220786204e-05,
      "loss": 0.1972,
      "num_input_tokens_seen": 16290176,
      "step": 10125
    },
    {
      "epoch": 0.4585889223386677,
      "grad_norm": 0.7979802489280701,
      "learning_rate": 4.249766157626409e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 16297952,
      "step": 10130
    },
    {
      "epoch": 0.45881527422531065,
      "grad_norm": 0.640255868434906,
      "learning_rate": 4.249064824630684e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 16306080,
      "step": 10135
    },
    {
      "epoch": 0.45904162611195365,
      "grad_norm": 0.8633829951286316,
      "learning_rate": 4.248363221907183e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 16314688,
      "step": 10140
    },
    {
      "epoch": 0.45926797799859664,
      "grad_norm": 0.3727962374687195,
      "learning_rate": 4.2476613495641026e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 16322752,
      "step": 10145
    },
    {
      "epoch": 0.4594943298852396,
      "grad_norm": 1.086321473121643,
      "learning_rate": 4.246959207709679e-05,
      "loss": 0.1813,
      "num_input_tokens_seen": 16331232,
      "step": 10150
    },
    {
      "epoch": 0.4597206817718826,
      "grad_norm": 0.6352584958076477,
      "learning_rate": 4.246256796452192e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 16339232,
      "step": 10155
    },
    {
      "epoch": 0.4599470336585255,
      "grad_norm": 0.7738851308822632,
      "learning_rate": 4.245554115899962e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 16347392,
      "step": 10160
    },
    {
      "epoch": 0.4601733855451685,
      "grad_norm": 1.2746821641921997,
      "learning_rate": 4.2448511661613514e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 16355808,
      "step": 10165
    },
    {
      "epoch": 0.4603997374318115,
      "grad_norm": 0.35144054889678955,
      "learning_rate": 4.2441479473447635e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 16364128,
      "step": 10170
    },
    {
      "epoch": 0.46062608931845445,
      "grad_norm": 0.6377061605453491,
      "learning_rate": 4.243444459558644e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 16372096,
      "step": 10175
    },
    {
      "epoch": 0.46085244120509744,
      "grad_norm": 1.1172202825546265,
      "learning_rate": 4.24274070291148e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 16379936,
      "step": 10180
    },
    {
      "epoch": 0.46107879309174044,
      "grad_norm": 0.9549902081489563,
      "learning_rate": 4.242036677511798e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 16388448,
      "step": 10185
    },
    {
      "epoch": 0.4613051449783834,
      "grad_norm": 0.46121788024902344,
      "learning_rate": 4.241332383468169e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 16396608,
      "step": 10190
    },
    {
      "epoch": 0.46153149686502637,
      "grad_norm": 0.5297004580497742,
      "learning_rate": 4.2406278208892034e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 16404128,
      "step": 10195
    },
    {
      "epoch": 0.46175784875166936,
      "grad_norm": 1.1052641868591309,
      "learning_rate": 4.2399229898835536e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 16413248,
      "step": 10200
    },
    {
      "epoch": 0.46175784875166936,
      "eval_loss": 0.15941131114959717,
      "eval_runtime": 405.0831,
      "eval_samples_per_second": 96.946,
      "eval_steps_per_second": 24.237,
      "num_input_tokens_seen": 16413248,
      "step": 10200
    },
    {
      "epoch": 0.4619842006383123,
      "grad_norm": 0.8637098670005798,
      "learning_rate": 4.239217890559914e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 16421632,
      "step": 10205
    },
    {
      "epoch": 0.4622105525249553,
      "grad_norm": 0.6450507044792175,
      "learning_rate": 4.238512523027019e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 16429888,
      "step": 10210
    },
    {
      "epoch": 0.4624369044115983,
      "grad_norm": 0.5459203720092773,
      "learning_rate": 4.237806887393645e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 16437952,
      "step": 10215
    },
    {
      "epoch": 0.46266325629824123,
      "grad_norm": 0.45601725578308105,
      "learning_rate": 4.237100983768611e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 16446080,
      "step": 10220
    },
    {
      "epoch": 0.4628896081848842,
      "grad_norm": 0.7144187688827515,
      "learning_rate": 4.2363948122607756e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 16454112,
      "step": 10225
    },
    {
      "epoch": 0.4631159600715272,
      "grad_norm": 0.5606911778450012,
      "learning_rate": 4.235688372979039e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 16462048,
      "step": 10230
    },
    {
      "epoch": 0.46334231195817016,
      "grad_norm": 0.4239911139011383,
      "learning_rate": 4.234981666032343e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 16470432,
      "step": 10235
    },
    {
      "epoch": 0.46356866384481316,
      "grad_norm": 0.3349953889846802,
      "learning_rate": 4.2342746915296704e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 16478272,
      "step": 10240
    },
    {
      "epoch": 0.4637950157314561,
      "grad_norm": 0.809150218963623,
      "learning_rate": 4.233567449580047e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 16486240,
      "step": 10245
    },
    {
      "epoch": 0.4640213676180991,
      "grad_norm": 0.8848648071289062,
      "learning_rate": 4.232859940292537e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 16494784,
      "step": 10250
    },
    {
      "epoch": 0.4642477195047421,
      "grad_norm": 0.5307698249816895,
      "learning_rate": 4.232152163776248e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 16502528,
      "step": 10255
    },
    {
      "epoch": 0.464474071391385,
      "grad_norm": 0.4149954617023468,
      "learning_rate": 4.231444120140328e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 16510720,
      "step": 10260
    },
    {
      "epoch": 0.464700423278028,
      "grad_norm": 0.6467314958572388,
      "learning_rate": 4.230735809493967e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 16518976,
      "step": 10265
    },
    {
      "epoch": 0.464926775164671,
      "grad_norm": 0.9667589068412781,
      "learning_rate": 4.2300272319463926e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 16526848,
      "step": 10270
    },
    {
      "epoch": 0.46515312705131395,
      "grad_norm": 0.6529710292816162,
      "learning_rate": 4.2293183876068786e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 16535296,
      "step": 10275
    },
    {
      "epoch": 0.46537947893795695,
      "grad_norm": 0.6127710938453674,
      "learning_rate": 4.228609276584737e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 16543648,
      "step": 10280
    },
    {
      "epoch": 0.46560583082459994,
      "grad_norm": 1.681374192237854,
      "learning_rate": 4.227899898989323e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 16552128,
      "step": 10285
    },
    {
      "epoch": 0.4658321827112429,
      "grad_norm": 0.5130391716957092,
      "learning_rate": 4.2271902549300293e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 16560032,
      "step": 10290
    },
    {
      "epoch": 0.4660585345978859,
      "grad_norm": 1.0971637964248657,
      "learning_rate": 4.226480344516294e-05,
      "loss": 0.2026,
      "num_input_tokens_seen": 16567744,
      "step": 10295
    },
    {
      "epoch": 0.46628488648452887,
      "grad_norm": 0.5647746324539185,
      "learning_rate": 4.2257701678575925e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 16575776,
      "step": 10300
    },
    {
      "epoch": 0.4665112383711718,
      "grad_norm": 0.8043989539146423,
      "learning_rate": 4.225059725063444e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 16583456,
      "step": 10305
    },
    {
      "epoch": 0.4667375902578148,
      "grad_norm": 0.7272403240203857,
      "learning_rate": 4.2243490162434074e-05,
      "loss": 0.1887,
      "num_input_tokens_seen": 16591008,
      "step": 10310
    },
    {
      "epoch": 0.4669639421444578,
      "grad_norm": 0.7796876430511475,
      "learning_rate": 4.223638041507083e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 16599584,
      "step": 10315
    },
    {
      "epoch": 0.46719029403110074,
      "grad_norm": 0.7292349338531494,
      "learning_rate": 4.2229268009641124e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 16608288,
      "step": 10320
    },
    {
      "epoch": 0.46741664591774373,
      "grad_norm": 1.037636637687683,
      "learning_rate": 4.222215294724177e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 16616640,
      "step": 10325
    },
    {
      "epoch": 0.4676429978043867,
      "grad_norm": 0.5546159148216248,
      "learning_rate": 4.2215035228970005e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 16624640,
      "step": 10330
    },
    {
      "epoch": 0.46786934969102967,
      "grad_norm": 0.7075117826461792,
      "learning_rate": 4.2207914855923464e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 16633088,
      "step": 10335
    },
    {
      "epoch": 0.46809570157767266,
      "grad_norm": 0.34749579429626465,
      "learning_rate": 4.220079182920021e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 16641280,
      "step": 10340
    },
    {
      "epoch": 0.4683220534643156,
      "grad_norm": 1.3342843055725098,
      "learning_rate": 4.2193666149898705e-05,
      "loss": 0.1746,
      "num_input_tokens_seen": 16650560,
      "step": 10345
    },
    {
      "epoch": 0.4685484053509586,
      "grad_norm": 0.8398526310920715,
      "learning_rate": 4.21865378191178e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 16658848,
      "step": 10350
    },
    {
      "epoch": 0.4687747572376016,
      "grad_norm": 0.8054515719413757,
      "learning_rate": 4.217940683795678e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 16667168,
      "step": 10355
    },
    {
      "epoch": 0.46900110912424453,
      "grad_norm": 0.9201962947845459,
      "learning_rate": 4.217227320751534e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 16674912,
      "step": 10360
    },
    {
      "epoch": 0.4692274610108875,
      "grad_norm": 0.27764609456062317,
      "learning_rate": 4.216513692889358e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 16682912,
      "step": 10365
    },
    {
      "epoch": 0.4694538128975305,
      "grad_norm": 0.7023159265518188,
      "learning_rate": 4.215799800319199e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 16691616,
      "step": 10370
    },
    {
      "epoch": 0.46968016478417346,
      "grad_norm": 0.4265337288379669,
      "learning_rate": 4.2150856431511485e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 16699616,
      "step": 10375
    },
    {
      "epoch": 0.46990651667081645,
      "grad_norm": 0.5614351630210876,
      "learning_rate": 4.214371221495339e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 16708512,
      "step": 10380
    },
    {
      "epoch": 0.47013286855745945,
      "grad_norm": 1.8878262042999268,
      "learning_rate": 4.213656535461942e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 16716608,
      "step": 10385
    },
    {
      "epoch": 0.4703592204441024,
      "grad_norm": 0.5178304314613342,
      "learning_rate": 4.2129415851611734e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 16724768,
      "step": 10390
    },
    {
      "epoch": 0.4705855723307454,
      "grad_norm": 1.0232946872711182,
      "learning_rate": 4.2122263707032855e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 16733472,
      "step": 10395
    },
    {
      "epoch": 0.4708119242173884,
      "grad_norm": 0.7435044050216675,
      "learning_rate": 4.211510892198574e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 16741440,
      "step": 10400
    },
    {
      "epoch": 0.4708119242173884,
      "eval_loss": 0.15880638360977173,
      "eval_runtime": 405.067,
      "eval_samples_per_second": 96.949,
      "eval_steps_per_second": 24.238,
      "num_input_tokens_seen": 16741440,
      "step": 10400
    },
    {
      "epoch": 0.4710382761040313,
      "grad_norm": 1.429991602897644,
      "learning_rate": 4.210795149757375e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 16749536,
      "step": 10405
    },
    {
      "epoch": 0.4712646279906743,
      "grad_norm": 0.9666211009025574,
      "learning_rate": 4.210079143490065e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 16757152,
      "step": 10410
    },
    {
      "epoch": 0.47149097987731725,
      "grad_norm": 1.428836464881897,
      "learning_rate": 4.2093628735070604e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 16765696,
      "step": 10415
    },
    {
      "epoch": 0.47171733176396025,
      "grad_norm": 1.333831548690796,
      "learning_rate": 4.208646339918819e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 16773856,
      "step": 10420
    },
    {
      "epoch": 0.47194368365060324,
      "grad_norm": 1.0576763153076172,
      "learning_rate": 4.2079295428358414e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 16780992,
      "step": 10425
    },
    {
      "epoch": 0.4721700355372462,
      "grad_norm": 0.8868674635887146,
      "learning_rate": 4.207212482368664e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 16789216,
      "step": 10430
    },
    {
      "epoch": 0.4723963874238892,
      "grad_norm": 0.36875441670417786,
      "learning_rate": 4.206495158627867e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 16796928,
      "step": 10435
    },
    {
      "epoch": 0.47262273931053217,
      "grad_norm": 0.5894221067428589,
      "learning_rate": 4.205777571724073e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 16804800,
      "step": 10440
    },
    {
      "epoch": 0.4728490911971751,
      "grad_norm": 0.8520549535751343,
      "learning_rate": 4.20505972176794e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 16812896,
      "step": 10445
    },
    {
      "epoch": 0.4730754430838181,
      "grad_norm": 0.8213192224502563,
      "learning_rate": 4.204341608870171e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 16820352,
      "step": 10450
    },
    {
      "epoch": 0.4733017949704611,
      "grad_norm": 1.4515223503112793,
      "learning_rate": 4.203623233141508e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 16828256,
      "step": 10455
    },
    {
      "epoch": 0.47352814685710404,
      "grad_norm": 0.5149920582771301,
      "learning_rate": 4.2029045946927334e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 16836320,
      "step": 10460
    },
    {
      "epoch": 0.47375449874374703,
      "grad_norm": 0.470831960439682,
      "learning_rate": 4.20218569363467e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 16844832,
      "step": 10465
    },
    {
      "epoch": 0.47398085063039,
      "grad_norm": 1.2441171407699585,
      "learning_rate": 4.2014665300781834e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 16852480,
      "step": 10470
    },
    {
      "epoch": 0.47420720251703297,
      "grad_norm": 1.1883904933929443,
      "learning_rate": 4.200747104134174e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 16860704,
      "step": 10475
    },
    {
      "epoch": 0.47443355440367596,
      "grad_norm": 0.4452524781227112,
      "learning_rate": 4.200027415913588e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 16868576,
      "step": 10480
    },
    {
      "epoch": 0.47465990629031896,
      "grad_norm": 0.7239815592765808,
      "learning_rate": 4.1993074655274126e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 16876864,
      "step": 10485
    },
    {
      "epoch": 0.4748862581769619,
      "grad_norm": 0.6376639604568481,
      "learning_rate": 4.198587253086669e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 16884416,
      "step": 10490
    },
    {
      "epoch": 0.4751126100636049,
      "grad_norm": 0.2176227867603302,
      "learning_rate": 4.197866778702426e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 16892512,
      "step": 10495
    },
    {
      "epoch": 0.47533896195024783,
      "grad_norm": 0.6003966331481934,
      "learning_rate": 4.197146042485789e-05,
      "loss": 0.1159,
      "num_input_tokens_seen": 16900320,
      "step": 10500
    },
    {
      "epoch": 0.4755653138368908,
      "grad_norm": 0.3743134140968323,
      "learning_rate": 4.1964250445479046e-05,
      "loss": 0.2039,
      "num_input_tokens_seen": 16909120,
      "step": 10505
    },
    {
      "epoch": 0.4757916657235338,
      "grad_norm": 0.716300368309021,
      "learning_rate": 4.19570378499996e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 16917440,
      "step": 10510
    },
    {
      "epoch": 0.47601801761017676,
      "grad_norm": 1.0460447072982788,
      "learning_rate": 4.194982263953182e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 16925248,
      "step": 10515
    },
    {
      "epoch": 0.47624436949681975,
      "grad_norm": 0.44387081265449524,
      "learning_rate": 4.194260481518838e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 16933024,
      "step": 10520
    },
    {
      "epoch": 0.47647072138346275,
      "grad_norm": 0.4240231513977051,
      "learning_rate": 4.1935384378082366e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 16941120,
      "step": 10525
    },
    {
      "epoch": 0.4766970732701057,
      "grad_norm": 0.6098174452781677,
      "learning_rate": 4.1928161329327267e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 16948544,
      "step": 10530
    },
    {
      "epoch": 0.4769234251567487,
      "grad_norm": 1.1812013387680054,
      "learning_rate": 4.1920935670036945e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 16955808,
      "step": 10535
    },
    {
      "epoch": 0.4771497770433917,
      "grad_norm": 0.5952178835868835,
      "learning_rate": 4.1913707401325705e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 16964000,
      "step": 10540
    },
    {
      "epoch": 0.4773761289300346,
      "grad_norm": 0.9200071096420288,
      "learning_rate": 4.1906476524308235e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 16971424,
      "step": 10545
    },
    {
      "epoch": 0.4776024808166776,
      "grad_norm": 0.95045006275177,
      "learning_rate": 4.189924304009962e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 16979552,
      "step": 10550
    },
    {
      "epoch": 0.4778288327033206,
      "grad_norm": 0.8041160702705383,
      "learning_rate": 4.189200694981537e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 16987616,
      "step": 10555
    },
    {
      "epoch": 0.47805518458996354,
      "grad_norm": 0.6117891073226929,
      "learning_rate": 4.188476825457136e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 16995616,
      "step": 10560
    },
    {
      "epoch": 0.47828153647660654,
      "grad_norm": 0.6335170865058899,
      "learning_rate": 4.18775269554839e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 17004096,
      "step": 10565
    },
    {
      "epoch": 0.47850788836324953,
      "grad_norm": 0.8972837924957275,
      "learning_rate": 4.187028305366969e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 17012928,
      "step": 10570
    },
    {
      "epoch": 0.4787342402498925,
      "grad_norm": 1.1445730924606323,
      "learning_rate": 4.1863036550245824e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 17020992,
      "step": 10575
    },
    {
      "epoch": 0.47896059213653547,
      "grad_norm": 1.2379355430603027,
      "learning_rate": 4.1855787446329806e-05,
      "loss": 0.1881,
      "num_input_tokens_seen": 17029120,
      "step": 10580
    },
    {
      "epoch": 0.4791869440231784,
      "grad_norm": 0.29525065422058105,
      "learning_rate": 4.184853574303955e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 17037280,
      "step": 10585
    },
    {
      "epoch": 0.4794132959098214,
      "grad_norm": 0.6066020131111145,
      "learning_rate": 4.184128144149334e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 17045984,
      "step": 10590
    },
    {
      "epoch": 0.4796396477964644,
      "grad_norm": 0.7351398468017578,
      "learning_rate": 4.1834024542809896e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 17053792,
      "step": 10595
    },
    {
      "epoch": 0.47986599968310734,
      "grad_norm": 1.2351016998291016,
      "learning_rate": 4.1826765048108315e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 17061536,
      "step": 10600
    },
    {
      "epoch": 0.47986599968310734,
      "eval_loss": 0.1583527773618698,
      "eval_runtime": 404.9838,
      "eval_samples_per_second": 96.969,
      "eval_steps_per_second": 24.243,
      "num_input_tokens_seen": 17061536,
      "step": 10600
    },
    {
      "epoch": 0.48009235156975033,
      "grad_norm": 0.662126362323761,
      "learning_rate": 4.181950295850811e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 17070336,
      "step": 10605
    },
    {
      "epoch": 0.4803187034563933,
      "grad_norm": 0.4648870527744293,
      "learning_rate": 4.181223827512918e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 17078624,
      "step": 10610
    },
    {
      "epoch": 0.48054505534303626,
      "grad_norm": 0.542806088924408,
      "learning_rate": 4.180497099909183e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 17086176,
      "step": 10615
    },
    {
      "epoch": 0.48077140722967926,
      "grad_norm": 0.7583444118499756,
      "learning_rate": 4.179770113151677e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 17094304,
      "step": 10620
    },
    {
      "epoch": 0.48099775911632225,
      "grad_norm": 0.37525510787963867,
      "learning_rate": 4.179042867352511e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 17102656,
      "step": 10625
    },
    {
      "epoch": 0.4812241110029652,
      "grad_norm": 0.6124613285064697,
      "learning_rate": 4.1783153626238334e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 17110592,
      "step": 10630
    },
    {
      "epoch": 0.4814504628896082,
      "grad_norm": 0.7500883936882019,
      "learning_rate": 4.177587599077836e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 17118464,
      "step": 10635
    },
    {
      "epoch": 0.4816768147762512,
      "grad_norm": 0.7716211080551147,
      "learning_rate": 4.1768595768267494e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 17126560,
      "step": 10640
    },
    {
      "epoch": 0.4819031666628941,
      "grad_norm": 0.6749695539474487,
      "learning_rate": 4.176131295982843e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 17134784,
      "step": 10645
    },
    {
      "epoch": 0.4821295185495371,
      "grad_norm": 0.4341510534286499,
      "learning_rate": 4.1754027566584276e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 17143296,
      "step": 10650
    },
    {
      "epoch": 0.4823558704361801,
      "grad_norm": 0.6463249921798706,
      "learning_rate": 4.174673958965852e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 17151232,
      "step": 10655
    },
    {
      "epoch": 0.48258222232282305,
      "grad_norm": 0.4327579438686371,
      "learning_rate": 4.173944903017507e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 17158912,
      "step": 10660
    },
    {
      "epoch": 0.48280857420946605,
      "grad_norm": 0.4827159345149994,
      "learning_rate": 4.173215588925822e-05,
      "loss": 0.1693,
      "num_input_tokens_seen": 17167040,
      "step": 10665
    },
    {
      "epoch": 0.483034926096109,
      "grad_norm": 0.6650387644767761,
      "learning_rate": 4.172486016803266e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 17174976,
      "step": 10670
    },
    {
      "epoch": 0.483261277982752,
      "grad_norm": 0.7715134024620056,
      "learning_rate": 4.171756186762349e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 17183488,
      "step": 10675
    },
    {
      "epoch": 0.483487629869395,
      "grad_norm": 0.4910060465335846,
      "learning_rate": 4.171026098915619e-05,
      "loss": 0.1026,
      "num_input_tokens_seen": 17190752,
      "step": 10680
    },
    {
      "epoch": 0.4837139817560379,
      "grad_norm": 0.995186984539032,
      "learning_rate": 4.170295753375665e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 17198880,
      "step": 10685
    },
    {
      "epoch": 0.4839403336426809,
      "grad_norm": 0.9107761383056641,
      "learning_rate": 4.169565150255117e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 17207424,
      "step": 10690
    },
    {
      "epoch": 0.4841666855293239,
      "grad_norm": 0.6742346882820129,
      "learning_rate": 4.16883428966664e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 17215040,
      "step": 10695
    },
    {
      "epoch": 0.48439303741596684,
      "grad_norm": 0.6013103127479553,
      "learning_rate": 4.168103171722944e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 17222944,
      "step": 10700
    },
    {
      "epoch": 0.48461938930260984,
      "grad_norm": 0.7175770401954651,
      "learning_rate": 4.167371796536777e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 17231648,
      "step": 10705
    },
    {
      "epoch": 0.48484574118925283,
      "grad_norm": 0.7628891468048096,
      "learning_rate": 4.166640164220924e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 17239680,
      "step": 10710
    },
    {
      "epoch": 0.48507209307589577,
      "grad_norm": 1.1612969636917114,
      "learning_rate": 4.1659082748882144e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 17247328,
      "step": 10715
    },
    {
      "epoch": 0.48529844496253877,
      "grad_norm": 0.5360262989997864,
      "learning_rate": 4.1651761286515135e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 17255584,
      "step": 10720
    },
    {
      "epoch": 0.48552479684918176,
      "grad_norm": 0.5146010518074036,
      "learning_rate": 4.164443725623728e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 17263936,
      "step": 10725
    },
    {
      "epoch": 0.4857511487358247,
      "grad_norm": 0.7607476711273193,
      "learning_rate": 4.163711065917802e-05,
      "loss": 0.1716,
      "num_input_tokens_seen": 17271168,
      "step": 10730
    },
    {
      "epoch": 0.4859775006224677,
      "grad_norm": 0.7831059694290161,
      "learning_rate": 4.1629781496467234e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 17279168,
      "step": 10735
    },
    {
      "epoch": 0.4862038525091107,
      "grad_norm": 0.5445549488067627,
      "learning_rate": 4.1622449769235164e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 17286656,
      "step": 10740
    },
    {
      "epoch": 0.48643020439575363,
      "grad_norm": 1.3119064569473267,
      "learning_rate": 4.161511547861243e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 17294592,
      "step": 10745
    },
    {
      "epoch": 0.4866565562823966,
      "grad_norm": 1.0467894077301025,
      "learning_rate": 4.1607778625730104e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 17302368,
      "step": 10750
    },
    {
      "epoch": 0.48688290816903956,
      "grad_norm": 0.6277452707290649,
      "learning_rate": 4.160043921171961e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 17310720,
      "step": 10755
    },
    {
      "epoch": 0.48710926005568256,
      "grad_norm": 0.5836737155914307,
      "learning_rate": 4.159309723771276e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 17318496,
      "step": 10760
    },
    {
      "epoch": 0.48733561194232555,
      "grad_norm": 0.5514728426933289,
      "learning_rate": 4.158575270484181e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 17326880,
      "step": 10765
    },
    {
      "epoch": 0.4875619638289685,
      "grad_norm": 0.45649346709251404,
      "learning_rate": 4.157840561423936e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 17334464,
      "step": 10770
    },
    {
      "epoch": 0.4877883157156115,
      "grad_norm": 0.5191328525543213,
      "learning_rate": 4.1571055967038416e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 17342368,
      "step": 10775
    },
    {
      "epoch": 0.4880146676022545,
      "grad_norm": 0.9967218637466431,
      "learning_rate": 4.156370376437241e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 17350848,
      "step": 10780
    },
    {
      "epoch": 0.4882410194888974,
      "grad_norm": 0.4773012697696686,
      "learning_rate": 4.155634900737513e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 17358848,
      "step": 10785
    },
    {
      "epoch": 0.4884673713755404,
      "grad_norm": 0.5351541042327881,
      "learning_rate": 4.1548991697180764e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 17367360,
      "step": 10790
    },
    {
      "epoch": 0.4886937232621834,
      "grad_norm": 1.0863780975341797,
      "learning_rate": 4.1541631834923914e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 17375552,
      "step": 10795
    },
    {
      "epoch": 0.48892007514882635,
      "grad_norm": 0.48417219519615173,
      "learning_rate": 4.153426942173956e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 17383360,
      "step": 10800
    },
    {
      "epoch": 0.48892007514882635,
      "eval_loss": 0.15758657455444336,
      "eval_runtime": 404.2103,
      "eval_samples_per_second": 97.155,
      "eval_steps_per_second": 24.289,
      "num_input_tokens_seen": 17383360,
      "step": 10800
    },
    {
      "epoch": 0.48914642703546934,
      "grad_norm": 0.5352222919464111,
      "learning_rate": 4.152690445876308e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 17391584,
      "step": 10805
    },
    {
      "epoch": 0.48937277892211234,
      "grad_norm": 0.4137296974658966,
      "learning_rate": 4.1519536947130245e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 17400256,
      "step": 10810
    },
    {
      "epoch": 0.4895991308087553,
      "grad_norm": 0.7446067333221436,
      "learning_rate": 4.151216688797722e-05,
      "loss": 0.1771,
      "num_input_tokens_seen": 17407744,
      "step": 10815
    },
    {
      "epoch": 0.4898254826953983,
      "grad_norm": 0.6368945837020874,
      "learning_rate": 4.150479428244054e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 17415776,
      "step": 10820
    },
    {
      "epoch": 0.49005183458204127,
      "grad_norm": 0.41306111216545105,
      "learning_rate": 4.1497419131657176e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 17423232,
      "step": 10825
    },
    {
      "epoch": 0.4902781864686842,
      "grad_norm": 0.6418704986572266,
      "learning_rate": 4.149004143676447e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 17431008,
      "step": 10830
    },
    {
      "epoch": 0.4905045383553272,
      "grad_norm": 0.4030733108520508,
      "learning_rate": 4.148266119890015e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 17439168,
      "step": 10835
    },
    {
      "epoch": 0.49073089024197014,
      "grad_norm": 1.0775595903396606,
      "learning_rate": 4.1475278419202324e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 17447424,
      "step": 10840
    },
    {
      "epoch": 0.49095724212861314,
      "grad_norm": 0.6622448563575745,
      "learning_rate": 4.146789309880953e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 17455584,
      "step": 10845
    },
    {
      "epoch": 0.49118359401525613,
      "grad_norm": 0.6207606792449951,
      "learning_rate": 4.146050523886068e-05,
      "loss": 0.2052,
      "num_input_tokens_seen": 17463648,
      "step": 10850
    },
    {
      "epoch": 0.49140994590189907,
      "grad_norm": 0.5983858108520508,
      "learning_rate": 4.1453114840495055e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 17471936,
      "step": 10855
    },
    {
      "epoch": 0.49163629778854206,
      "grad_norm": 0.71467524766922,
      "learning_rate": 4.1445721904852364e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 17480192,
      "step": 10860
    },
    {
      "epoch": 0.49186264967518506,
      "grad_norm": 0.7139292359352112,
      "learning_rate": 4.143832643307269e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 17489344,
      "step": 10865
    },
    {
      "epoch": 0.492089001561828,
      "grad_norm": 0.9245090484619141,
      "learning_rate": 4.1430928426296503e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 17497216,
      "step": 10870
    },
    {
      "epoch": 0.492315353448471,
      "grad_norm": 0.917694628238678,
      "learning_rate": 4.142352788566466e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 17504800,
      "step": 10875
    },
    {
      "epoch": 0.492541705335114,
      "grad_norm": 0.7076456546783447,
      "learning_rate": 4.1416124812318424e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 17512768,
      "step": 10880
    },
    {
      "epoch": 0.4927680572217569,
      "grad_norm": 0.49871817231178284,
      "learning_rate": 4.1408719207399453e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 17521792,
      "step": 10885
    },
    {
      "epoch": 0.4929944091083999,
      "grad_norm": 0.43383416533470154,
      "learning_rate": 4.140131107204978e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 17530240,
      "step": 10890
    },
    {
      "epoch": 0.4932207609950429,
      "grad_norm": 0.5391258001327515,
      "learning_rate": 4.139390040741182e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 17538080,
      "step": 10895
    },
    {
      "epoch": 0.49344711288168586,
      "grad_norm": 0.49241283535957336,
      "learning_rate": 4.1386487214628396e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 17545792,
      "step": 10900
    },
    {
      "epoch": 0.49367346476832885,
      "grad_norm": 0.8937309980392456,
      "learning_rate": 4.137907149484272e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 17554592,
      "step": 10905
    },
    {
      "epoch": 0.49389981665497185,
      "grad_norm": 0.40318989753723145,
      "learning_rate": 4.137165324919839e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 17562304,
      "step": 10910
    },
    {
      "epoch": 0.4941261685416148,
      "grad_norm": 0.6117059588432312,
      "learning_rate": 4.136423247883939e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 17570080,
      "step": 10915
    },
    {
      "epoch": 0.4943525204282578,
      "grad_norm": 0.9075640439987183,
      "learning_rate": 4.135680918491009e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 17577760,
      "step": 10920
    },
    {
      "epoch": 0.4945788723149007,
      "grad_norm": 0.39861956238746643,
      "learning_rate": 4.1349383368555265e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 17586240,
      "step": 10925
    },
    {
      "epoch": 0.4948052242015437,
      "grad_norm": 0.5976719856262207,
      "learning_rate": 4.1341955030920065e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 17594240,
      "step": 10930
    },
    {
      "epoch": 0.4950315760881867,
      "grad_norm": 0.5899489521980286,
      "learning_rate": 4.1334524173150036e-05,
      "loss": 0.1499,
      "num_input_tokens_seen": 17602784,
      "step": 10935
    },
    {
      "epoch": 0.49525792797482965,
      "grad_norm": 0.7059568166732788,
      "learning_rate": 4.13270907963911e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 17610560,
      "step": 10940
    },
    {
      "epoch": 0.49548427986147264,
      "grad_norm": 0.6565309762954712,
      "learning_rate": 4.131965490178959e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 17618816,
      "step": 10945
    },
    {
      "epoch": 0.49571063174811564,
      "grad_norm": 0.5855212807655334,
      "learning_rate": 4.131221649049222e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 17626720,
      "step": 10950
    },
    {
      "epoch": 0.4959369836347586,
      "grad_norm": 0.6767222881317139,
      "learning_rate": 4.130477556364606e-05,
      "loss": 0.1934,
      "num_input_tokens_seen": 17634592,
      "step": 10955
    },
    {
      "epoch": 0.49616333552140157,
      "grad_norm": 0.6728866696357727,
      "learning_rate": 4.129733212239861e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 17642560,
      "step": 10960
    },
    {
      "epoch": 0.49638968740804457,
      "grad_norm": 0.965705394744873,
      "learning_rate": 4.128988616789774e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 17651616,
      "step": 10965
    },
    {
      "epoch": 0.4966160392946875,
      "grad_norm": 0.7163441181182861,
      "learning_rate": 4.1282437701291724e-05,
      "loss": 0.1924,
      "num_input_tokens_seen": 17659168,
      "step": 10970
    },
    {
      "epoch": 0.4968423911813305,
      "grad_norm": 0.8624898195266724,
      "learning_rate": 4.1274986723729184e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 17666912,
      "step": 10975
    },
    {
      "epoch": 0.4970687430679735,
      "grad_norm": 0.841744601726532,
      "learning_rate": 4.126753323635917e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 17675296,
      "step": 10980
    },
    {
      "epoch": 0.49729509495461643,
      "grad_norm": 0.477746844291687,
      "learning_rate": 4.12600772403311e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 17684224,
      "step": 10985
    },
    {
      "epoch": 0.49752144684125943,
      "grad_norm": 0.4701358675956726,
      "learning_rate": 4.125261873679479e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 17691936,
      "step": 10990
    },
    {
      "epoch": 0.4977477987279024,
      "grad_norm": 0.3786928057670593,
      "learning_rate": 4.124515772690042e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 17700288,
      "step": 10995
    },
    {
      "epoch": 0.49797415061454536,
      "grad_norm": 0.8678790330886841,
      "learning_rate": 4.123769421179858e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 17708608,
      "step": 11000
    },
    {
      "epoch": 0.49797415061454536,
      "eval_loss": 0.15706636011600494,
      "eval_runtime": 405.0588,
      "eval_samples_per_second": 96.951,
      "eval_steps_per_second": 24.238,
      "num_input_tokens_seen": 17708608,
      "step": 11000
    },
    {
      "epoch": 0.49820050250118836,
      "grad_norm": 1.2591317892074585,
      "learning_rate": 4.1230228192640236e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 17717056,
      "step": 11005
    },
    {
      "epoch": 0.4984268543878313,
      "grad_norm": 0.44015026092529297,
      "learning_rate": 4.122275967057675e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 17725280,
      "step": 11010
    },
    {
      "epoch": 0.4986532062744743,
      "grad_norm": 0.4362776577472687,
      "learning_rate": 4.1215288646759846e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 17732928,
      "step": 11015
    },
    {
      "epoch": 0.4988795581611173,
      "grad_norm": 0.7061010599136353,
      "learning_rate": 4.120781512234166e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 17740736,
      "step": 11020
    },
    {
      "epoch": 0.4991059100477602,
      "grad_norm": 0.46445509791374207,
      "learning_rate": 4.120033909847471e-05,
      "loss": 0.1012,
      "num_input_tokens_seen": 17748384,
      "step": 11025
    },
    {
      "epoch": 0.4993322619344032,
      "grad_norm": 0.6386488676071167,
      "learning_rate": 4.119286057631187e-05,
      "loss": 0.1821,
      "num_input_tokens_seen": 17755872,
      "step": 11030
    },
    {
      "epoch": 0.4995586138210462,
      "grad_norm": 1.2133476734161377,
      "learning_rate": 4.118537955700646e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 17764224,
      "step": 11035
    },
    {
      "epoch": 0.49978496570768915,
      "grad_norm": 0.6852270364761353,
      "learning_rate": 4.11778960417121e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 17772256,
      "step": 11040
    },
    {
      "epoch": 0.5000113175943321,
      "grad_norm": 0.5147890448570251,
      "learning_rate": 4.117041003158288e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 17780288,
      "step": 11045
    },
    {
      "epoch": 0.5002376694809751,
      "grad_norm": 0.6437789797782898,
      "learning_rate": 4.1162921527773215e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 17787936,
      "step": 11050
    },
    {
      "epoch": 0.5004640213676181,
      "grad_norm": 0.9964373111724854,
      "learning_rate": 4.115543053143794e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 17795712,
      "step": 11055
    },
    {
      "epoch": 0.500690373254261,
      "grad_norm": 0.8050958514213562,
      "learning_rate": 4.114793704373226e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 17804064,
      "step": 11060
    },
    {
      "epoch": 0.500916725140904,
      "grad_norm": 0.6142140030860901,
      "learning_rate": 4.114044106581175e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 17812992,
      "step": 11065
    },
    {
      "epoch": 0.501143077027547,
      "grad_norm": 1.1726442575454712,
      "learning_rate": 4.11329425988324e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 17821376,
      "step": 11070
    },
    {
      "epoch": 0.50136942891419,
      "grad_norm": 0.44104719161987305,
      "learning_rate": 4.112544164395056e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 17830048,
      "step": 11075
    },
    {
      "epoch": 0.501595780800833,
      "grad_norm": 0.688140869140625,
      "learning_rate": 4.111793820232297e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 17838112,
      "step": 11080
    },
    {
      "epoch": 0.501822132687476,
      "grad_norm": 1.0449806451797485,
      "learning_rate": 4.1110432275106767e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 17845824,
      "step": 11085
    },
    {
      "epoch": 0.5020484845741189,
      "grad_norm": 0.4726642072200775,
      "learning_rate": 4.110292386345944e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 17854016,
      "step": 11090
    },
    {
      "epoch": 0.5022748364607619,
      "grad_norm": 0.49961990118026733,
      "learning_rate": 4.109541296853891e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 17862624,
      "step": 11095
    },
    {
      "epoch": 0.5025011883474049,
      "grad_norm": 1.186229944229126,
      "learning_rate": 4.108789959150341e-05,
      "loss": 0.2033,
      "num_input_tokens_seen": 17871136,
      "step": 11100
    },
    {
      "epoch": 0.5027275402340479,
      "grad_norm": 0.39109697937965393,
      "learning_rate": 4.108038373351163e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 17879680,
      "step": 11105
    },
    {
      "epoch": 0.5029538921206909,
      "grad_norm": 0.35338592529296875,
      "learning_rate": 4.10728653957226e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 17887328,
      "step": 11110
    },
    {
      "epoch": 0.5031802440073339,
      "grad_norm": 0.567435622215271,
      "learning_rate": 4.106534457929575e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 17895264,
      "step": 11115
    },
    {
      "epoch": 0.5034065958939767,
      "grad_norm": 0.34858188033103943,
      "learning_rate": 4.105782128539086e-05,
      "loss": 0.0999,
      "num_input_tokens_seen": 17903488,
      "step": 11120
    },
    {
      "epoch": 0.5036329477806197,
      "grad_norm": 0.6024525165557861,
      "learning_rate": 4.1050295515168144e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 17912096,
      "step": 11125
    },
    {
      "epoch": 0.5038592996672627,
      "grad_norm": 0.6464850902557373,
      "learning_rate": 4.1042767269788155e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 17920224,
      "step": 11130
    },
    {
      "epoch": 0.5040856515539057,
      "grad_norm": 0.7591211199760437,
      "learning_rate": 4.103523655041185e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 17928352,
      "step": 11135
    },
    {
      "epoch": 0.5043120034405487,
      "grad_norm": 0.810623049736023,
      "learning_rate": 4.102770335820055e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 17936768,
      "step": 11140
    },
    {
      "epoch": 0.5045383553271916,
      "grad_norm": 0.8131706118583679,
      "learning_rate": 4.1020167694315984e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 17944704,
      "step": 11145
    },
    {
      "epoch": 0.5047647072138346,
      "grad_norm": 0.38972383737564087,
      "learning_rate": 4.101262955992023e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 17953088,
      "step": 11150
    },
    {
      "epoch": 0.5049910591004776,
      "grad_norm": 1.273190975189209,
      "learning_rate": 4.100508895617578e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 17960832,
      "step": 11155
    },
    {
      "epoch": 0.5052174109871206,
      "grad_norm": 0.41338202357292175,
      "learning_rate": 4.099754588424547e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 17968160,
      "step": 11160
    },
    {
      "epoch": 0.5054437628737636,
      "grad_norm": 0.4941604435443878,
      "learning_rate": 4.0990000345292546e-05,
      "loss": 0.2018,
      "num_input_tokens_seen": 17976128,
      "step": 11165
    },
    {
      "epoch": 0.5056701147604066,
      "grad_norm": 0.36254817247390747,
      "learning_rate": 4.098245234048064e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 17983840,
      "step": 11170
    },
    {
      "epoch": 0.5058964666470495,
      "grad_norm": 0.3627485930919647,
      "learning_rate": 4.0974901870973726e-05,
      "loss": 0.1735,
      "num_input_tokens_seen": 17991488,
      "step": 11175
    },
    {
      "epoch": 0.5061228185336925,
      "grad_norm": 1.1379927396774292,
      "learning_rate": 4.096734893793619e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 17999616,
      "step": 11180
    },
    {
      "epoch": 0.5063491704203354,
      "grad_norm": 0.912703812122345,
      "learning_rate": 4.095979354253279e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 18007584,
      "step": 11185
    },
    {
      "epoch": 0.5065755223069784,
      "grad_norm": 0.8758314847946167,
      "learning_rate": 4.0952235685928656e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 18015744,
      "step": 11190
    },
    {
      "epoch": 0.5068018741936214,
      "grad_norm": 1.3123077154159546,
      "learning_rate": 4.094467536928932e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 18023808,
      "step": 11195
    },
    {
      "epoch": 0.5070282260802644,
      "grad_norm": 0.8428137302398682,
      "learning_rate": 4.093711259378067e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 18032288,
      "step": 11200
    },
    {
      "epoch": 0.5070282260802644,
      "eval_loss": 0.15675213932991028,
      "eval_runtime": 404.2667,
      "eval_samples_per_second": 97.141,
      "eval_steps_per_second": 24.286,
      "num_input_tokens_seen": 18032288,
      "step": 11200
    },
    {
      "epoch": 0.5072545779669073,
      "grad_norm": 1.0547664165496826,
      "learning_rate": 4.092954736056897e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 18040672,
      "step": 11205
    },
    {
      "epoch": 0.5074809298535503,
      "grad_norm": 0.384956955909729,
      "learning_rate": 4.09219796708209e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 18048224,
      "step": 11210
    },
    {
      "epoch": 0.5077072817401933,
      "grad_norm": 0.7805509567260742,
      "learning_rate": 4.0914409525703464e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 18056256,
      "step": 11215
    },
    {
      "epoch": 0.5079336336268363,
      "grad_norm": 0.3910037577152252,
      "learning_rate": 4.090683692638408e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 18064160,
      "step": 11220
    },
    {
      "epoch": 0.5081599855134793,
      "grad_norm": 0.9645024538040161,
      "learning_rate": 4.089926187403056e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 18071904,
      "step": 11225
    },
    {
      "epoch": 0.5083863374001222,
      "grad_norm": 0.7994396090507507,
      "learning_rate": 4.0891684369811044e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 18080352,
      "step": 11230
    },
    {
      "epoch": 0.5086126892867652,
      "grad_norm": 0.6471363306045532,
      "learning_rate": 4.0884104414894107e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 18088128,
      "step": 11235
    },
    {
      "epoch": 0.5088390411734082,
      "grad_norm": 0.5799772143363953,
      "learning_rate": 4.087652201044864e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 18096064,
      "step": 11240
    },
    {
      "epoch": 0.5090653930600512,
      "grad_norm": 0.8277880549430847,
      "learning_rate": 4.086893715764397e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 18104000,
      "step": 11245
    },
    {
      "epoch": 0.5092917449466942,
      "grad_norm": 0.4419725239276886,
      "learning_rate": 4.086134985764977e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 18112288,
      "step": 11250
    },
    {
      "epoch": 0.5095180968333372,
      "grad_norm": 0.7096128463745117,
      "learning_rate": 4.0853760111636085e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 18121024,
      "step": 11255
    },
    {
      "epoch": 0.50974444871998,
      "grad_norm": 0.5247617959976196,
      "learning_rate": 4.084616792077337e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 18129504,
      "step": 11260
    },
    {
      "epoch": 0.509970800606623,
      "grad_norm": 0.8846770524978638,
      "learning_rate": 4.083857328623243e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 18137120,
      "step": 11265
    },
    {
      "epoch": 0.510197152493266,
      "grad_norm": 0.642780065536499,
      "learning_rate": 4.083097620918444e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 18145376,
      "step": 11270
    },
    {
      "epoch": 0.510423504379909,
      "grad_norm": 0.6090551614761353,
      "learning_rate": 4.082337669080097e-05,
      "loss": 0.2051,
      "num_input_tokens_seen": 18153344,
      "step": 11275
    },
    {
      "epoch": 0.510649856266552,
      "grad_norm": 0.5863849520683289,
      "learning_rate": 4.081577473225398e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 18161184,
      "step": 11280
    },
    {
      "epoch": 0.510876208153195,
      "grad_norm": 0.3551216423511505,
      "learning_rate": 4.080817033471577e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 18169216,
      "step": 11285
    },
    {
      "epoch": 0.5111025600398379,
      "grad_norm": 0.5767734050750732,
      "learning_rate": 4.080056349935903e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 18177344,
      "step": 11290
    },
    {
      "epoch": 0.5113289119264809,
      "grad_norm": 0.7698684930801392,
      "learning_rate": 4.079295422735684e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 18185376,
      "step": 11295
    },
    {
      "epoch": 0.5115552638131239,
      "grad_norm": 0.8319222331047058,
      "learning_rate": 4.078534251988264e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 18193056,
      "step": 11300
    },
    {
      "epoch": 0.5117816156997669,
      "grad_norm": 0.702690839767456,
      "learning_rate": 4.077772837811025e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 18201408,
      "step": 11305
    },
    {
      "epoch": 0.5120079675864099,
      "grad_norm": 0.29689303040504456,
      "learning_rate": 4.0770111803213874e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 18209696,
      "step": 11310
    },
    {
      "epoch": 0.5122343194730528,
      "grad_norm": 1.2265487909317017,
      "learning_rate": 4.076249279636807e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 18217952,
      "step": 11315
    },
    {
      "epoch": 0.5124606713596958,
      "grad_norm": 0.6253088116645813,
      "learning_rate": 4.075487135874781e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 18225600,
      "step": 11320
    },
    {
      "epoch": 0.5126870232463387,
      "grad_norm": 0.514245867729187,
      "learning_rate": 4.074724749152837e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 18233792,
      "step": 11325
    },
    {
      "epoch": 0.5129133751329817,
      "grad_norm": 0.5648496150970459,
      "learning_rate": 4.07396211958855e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 18242304,
      "step": 11330
    },
    {
      "epoch": 0.5131397270196247,
      "grad_norm": 0.5152528285980225,
      "learning_rate": 4.073199247299523e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 18249856,
      "step": 11335
    },
    {
      "epoch": 0.5133660789062677,
      "grad_norm": 0.5888036489486694,
      "learning_rate": 4.072436132403403e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 18257504,
      "step": 11340
    },
    {
      "epoch": 0.5135924307929106,
      "grad_norm": 0.5867425203323364,
      "learning_rate": 4.0716727750178704e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 18265376,
      "step": 11345
    },
    {
      "epoch": 0.5138187826795536,
      "grad_norm": 0.5856679677963257,
      "learning_rate": 4.0709091752606455e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 18273664,
      "step": 11350
    },
    {
      "epoch": 0.5140451345661966,
      "grad_norm": 0.36608365178108215,
      "learning_rate": 4.070145333249484e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 18281952,
      "step": 11355
    },
    {
      "epoch": 0.5142714864528396,
      "grad_norm": 1.418925166130066,
      "learning_rate": 4.069381249102181e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 18289856,
      "step": 11360
    },
    {
      "epoch": 0.5144978383394826,
      "grad_norm": 0.6709374189376831,
      "learning_rate": 4.0686169229365665e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 18297376,
      "step": 11365
    },
    {
      "epoch": 0.5147241902261256,
      "grad_norm": 0.5457881689071655,
      "learning_rate": 4.067852354870511e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 18305280,
      "step": 11370
    },
    {
      "epoch": 0.5149505421127685,
      "grad_norm": 0.730980634689331,
      "learning_rate": 4.067087545021919e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 18313152,
      "step": 11375
    },
    {
      "epoch": 0.5151768939994115,
      "grad_norm": 0.6022354364395142,
      "learning_rate": 4.066322493508734e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 18320704,
      "step": 11380
    },
    {
      "epoch": 0.5154032458860545,
      "grad_norm": 1.662070631980896,
      "learning_rate": 4.065557200448937e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 18328928,
      "step": 11385
    },
    {
      "epoch": 0.5156295977726975,
      "grad_norm": 0.7656280994415283,
      "learning_rate": 4.064791665960546e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 18336800,
      "step": 11390
    },
    {
      "epoch": 0.5158559496593405,
      "grad_norm": 0.3296193778514862,
      "learning_rate": 4.064025890161615e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 18345408,
      "step": 11395
    },
    {
      "epoch": 0.5160823015459833,
      "grad_norm": 0.43039581179618835,
      "learning_rate": 4.0632598731702373e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 18353024,
      "step": 11400
    },
    {
      "epoch": 0.5160823015459833,
      "eval_loss": 0.1573229879140854,
      "eval_runtime": 405.1862,
      "eval_samples_per_second": 96.921,
      "eval_steps_per_second": 24.231,
      "num_input_tokens_seen": 18353024,
      "step": 11400
    },
    {
      "epoch": 0.5163086534326263,
      "grad_norm": 0.4228028655052185,
      "learning_rate": 4.0624936151045426e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 18360608,
      "step": 11405
    },
    {
      "epoch": 0.5165350053192693,
      "grad_norm": 0.2880484163761139,
      "learning_rate": 4.061727116082696e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 18368480,
      "step": 11410
    },
    {
      "epoch": 0.5167613572059123,
      "grad_norm": 1.183347225189209,
      "learning_rate": 4.060960376222903e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 18376544,
      "step": 11415
    },
    {
      "epoch": 0.5169877090925553,
      "grad_norm": 1.194609522819519,
      "learning_rate": 4.0601933956434034e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 18384544,
      "step": 11420
    },
    {
      "epoch": 0.5172140609791983,
      "grad_norm": 0.4304652512073517,
      "learning_rate": 4.059426174462476e-05,
      "loss": 0.1192,
      "num_input_tokens_seen": 18393152,
      "step": 11425
    },
    {
      "epoch": 0.5174404128658412,
      "grad_norm": 0.890617847442627,
      "learning_rate": 4.058658712798435e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 18401344,
      "step": 11430
    },
    {
      "epoch": 0.5176667647524842,
      "grad_norm": 1.210352897644043,
      "learning_rate": 4.0578910107696336e-05,
      "loss": 0.1815,
      "num_input_tokens_seen": 18409664,
      "step": 11435
    },
    {
      "epoch": 0.5178931166391272,
      "grad_norm": 0.5577004551887512,
      "learning_rate": 4.05712306849446e-05,
      "loss": 0.1604,
      "num_input_tokens_seen": 18418176,
      "step": 11440
    },
    {
      "epoch": 0.5181194685257702,
      "grad_norm": 0.9338241219520569,
      "learning_rate": 4.0563548860913415e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 18426368,
      "step": 11445
    },
    {
      "epoch": 0.5183458204124132,
      "grad_norm": 0.91164630651474,
      "learning_rate": 4.0555864636787414e-05,
      "loss": 0.1956,
      "num_input_tokens_seen": 18434592,
      "step": 11450
    },
    {
      "epoch": 0.5185721722990562,
      "grad_norm": 0.3337092697620392,
      "learning_rate": 4.054817801375159e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 18443520,
      "step": 11455
    },
    {
      "epoch": 0.518798524185699,
      "grad_norm": 0.6058472990989685,
      "learning_rate": 4.054048899299134e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 18451424,
      "step": 11460
    },
    {
      "epoch": 0.519024876072342,
      "grad_norm": 0.5653842091560364,
      "learning_rate": 4.0532797575692385e-05,
      "loss": 0.2133,
      "num_input_tokens_seen": 18459552,
      "step": 11465
    },
    {
      "epoch": 0.519251227958985,
      "grad_norm": 0.7586373090744019,
      "learning_rate": 4.052510376304085e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 18467552,
      "step": 11470
    },
    {
      "epoch": 0.519477579845628,
      "grad_norm": 0.42488759756088257,
      "learning_rate": 4.051740755622321e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 18475392,
      "step": 11475
    },
    {
      "epoch": 0.519703931732271,
      "grad_norm": 0.9454230666160583,
      "learning_rate": 4.050970895642632e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 18483808,
      "step": 11480
    },
    {
      "epoch": 0.5199302836189139,
      "grad_norm": 0.3168727159500122,
      "learning_rate": 4.050200796483741e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 18492064,
      "step": 11485
    },
    {
      "epoch": 0.5201566355055569,
      "grad_norm": 0.7885881662368774,
      "learning_rate": 4.049430458264405e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 18500640,
      "step": 11490
    },
    {
      "epoch": 0.5203829873921999,
      "grad_norm": 0.6215025782585144,
      "learning_rate": 4.048659881103422e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 18508416,
      "step": 11495
    },
    {
      "epoch": 0.5206093392788429,
      "grad_norm": 0.38163790106773376,
      "learning_rate": 4.0478890651196235e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 18516000,
      "step": 11500
    },
    {
      "epoch": 0.5208356911654859,
      "grad_norm": 0.5107771754264832,
      "learning_rate": 4.047118010431879e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 18524224,
      "step": 11505
    },
    {
      "epoch": 0.5210620430521289,
      "grad_norm": 0.5558602213859558,
      "learning_rate": 4.046346717159094e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 18532288,
      "step": 11510
    },
    {
      "epoch": 0.5212883949387718,
      "grad_norm": 0.48576468229293823,
      "learning_rate": 4.045575185420214e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 18540448,
      "step": 11515
    },
    {
      "epoch": 0.5215147468254148,
      "grad_norm": 0.4454581141471863,
      "learning_rate": 4.0448034153342165e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 18548352,
      "step": 11520
    },
    {
      "epoch": 0.5217410987120578,
      "grad_norm": 0.6885257363319397,
      "learning_rate": 4.0440314070201194e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 18555904,
      "step": 11525
    },
    {
      "epoch": 0.5219674505987008,
      "grad_norm": 0.41420283913612366,
      "learning_rate": 4.043259160596976e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 18563360,
      "step": 11530
    },
    {
      "epoch": 0.5221938024853437,
      "grad_norm": 0.8326783776283264,
      "learning_rate": 4.0424866761838767e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 18571488,
      "step": 11535
    },
    {
      "epoch": 0.5224201543719867,
      "grad_norm": 1.134443759918213,
      "learning_rate": 4.041713953899948e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 18578848,
      "step": 11540
    },
    {
      "epoch": 0.5226465062586296,
      "grad_norm": 0.4087511897087097,
      "learning_rate": 4.0409409938643515e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 18586720,
      "step": 11545
    },
    {
      "epoch": 0.5228728581452726,
      "grad_norm": 0.49399876594543457,
      "learning_rate": 4.0401677961962904e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 18594496,
      "step": 11550
    },
    {
      "epoch": 0.5230992100319156,
      "grad_norm": 1.0132923126220703,
      "learning_rate": 4.039394361015001e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 18603296,
      "step": 11555
    },
    {
      "epoch": 0.5233255619185586,
      "grad_norm": 0.5471352934837341,
      "learning_rate": 4.038620688439755e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 18611136,
      "step": 11560
    },
    {
      "epoch": 0.5235519138052016,
      "grad_norm": 0.5351139307022095,
      "learning_rate": 4.037846778589862e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 18619616,
      "step": 11565
    },
    {
      "epoch": 0.5237782656918445,
      "grad_norm": 0.48638585209846497,
      "learning_rate": 4.0370726315846715e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 18627616,
      "step": 11570
    },
    {
      "epoch": 0.5240046175784875,
      "grad_norm": 0.7313265204429626,
      "learning_rate": 4.036298247543565e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 18635744,
      "step": 11575
    },
    {
      "epoch": 0.5242309694651305,
      "grad_norm": 1.2414888143539429,
      "learning_rate": 4.035523626585962e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 18644288,
      "step": 11580
    },
    {
      "epoch": 0.5244573213517735,
      "grad_norm": 1.2089176177978516,
      "learning_rate": 4.0347487688313194e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 18652608,
      "step": 11585
    },
    {
      "epoch": 0.5246836732384165,
      "grad_norm": 0.5316031575202942,
      "learning_rate": 4.0339736743991296e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 18660096,
      "step": 11590
    },
    {
      "epoch": 0.5249100251250595,
      "grad_norm": 1.0730164051055908,
      "learning_rate": 4.0331983434089227e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 18667904,
      "step": 11595
    },
    {
      "epoch": 0.5251363770117023,
      "grad_norm": 0.866611123085022,
      "learning_rate": 4.032422775980264e-05,
      "loss": 0.1788,
      "num_input_tokens_seen": 18677184,
      "step": 11600
    },
    {
      "epoch": 0.5251363770117023,
      "eval_loss": 0.1565578579902649,
      "eval_runtime": 404.8415,
      "eval_samples_per_second": 97.003,
      "eval_steps_per_second": 24.251,
      "num_input_tokens_seen": 18677184,
      "step": 11600
    },
    {
      "epoch": 0.5253627288983453,
      "grad_norm": 0.6069252490997314,
      "learning_rate": 4.031646972232754e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 18685184,
      "step": 11605
    },
    {
      "epoch": 0.5255890807849883,
      "grad_norm": 1.1101040840148926,
      "learning_rate": 4.0308709322860344e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 18693152,
      "step": 11610
    },
    {
      "epoch": 0.5258154326716313,
      "grad_norm": 0.3290245532989502,
      "learning_rate": 4.0300946562597784e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 18701024,
      "step": 11615
    },
    {
      "epoch": 0.5260417845582743,
      "grad_norm": 0.3997773826122284,
      "learning_rate": 4.029318144273698e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 18709152,
      "step": 11620
    },
    {
      "epoch": 0.5262681364449173,
      "grad_norm": 0.40677663683891296,
      "learning_rate": 4.0285413964475415e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 18717024,
      "step": 11625
    },
    {
      "epoch": 0.5264944883315602,
      "grad_norm": 0.47543272376060486,
      "learning_rate": 4.0277644129010927e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 18724928,
      "step": 11630
    },
    {
      "epoch": 0.5267208402182032,
      "grad_norm": 0.5537609457969666,
      "learning_rate": 4.0269871937541724e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 18733344,
      "step": 11635
    },
    {
      "epoch": 0.5269471921048462,
      "grad_norm": 1.4559450149536133,
      "learning_rate": 4.026209739126637e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 18741312,
      "step": 11640
    },
    {
      "epoch": 0.5271735439914892,
      "grad_norm": 0.6847120523452759,
      "learning_rate": 4.025432049138381e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 18749280,
      "step": 11645
    },
    {
      "epoch": 0.5273998958781322,
      "grad_norm": 0.6905028223991394,
      "learning_rate": 4.0246541239093325e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 18757280,
      "step": 11650
    },
    {
      "epoch": 0.5276262477647751,
      "grad_norm": 0.8496614694595337,
      "learning_rate": 4.023875963559459e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 18764736,
      "step": 11655
    },
    {
      "epoch": 0.5278525996514181,
      "grad_norm": 0.4913751184940338,
      "learning_rate": 4.023097568208761e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 18773120,
      "step": 11660
    },
    {
      "epoch": 0.5280789515380611,
      "grad_norm": 1.0376739501953125,
      "learning_rate": 4.022318937977277e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 18780640,
      "step": 11665
    },
    {
      "epoch": 0.528305303424704,
      "grad_norm": 0.3601890504360199,
      "learning_rate": 4.021540072985084e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 18789216,
      "step": 11670
    },
    {
      "epoch": 0.528531655311347,
      "grad_norm": 0.43155214190483093,
      "learning_rate": 4.020760973352289e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 18796928,
      "step": 11675
    },
    {
      "epoch": 0.52875800719799,
      "grad_norm": 0.8487978577613831,
      "learning_rate": 4.019981639199042e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 18804800,
      "step": 11680
    },
    {
      "epoch": 0.5289843590846329,
      "grad_norm": 0.4756297767162323,
      "learning_rate": 4.0192020706455245e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 18812736,
      "step": 11685
    },
    {
      "epoch": 0.5292107109712759,
      "grad_norm": 0.8849201798439026,
      "learning_rate": 4.018422267811956e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 18820672,
      "step": 11690
    },
    {
      "epoch": 0.5294370628579189,
      "grad_norm": 0.6006022691726685,
      "learning_rate": 4.017642230818592e-05,
      "loss": 0.1955,
      "num_input_tokens_seen": 18829280,
      "step": 11695
    },
    {
      "epoch": 0.5296634147445619,
      "grad_norm": 0.6723225712776184,
      "learning_rate": 4.0168619597857246e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 18837824,
      "step": 11700
    },
    {
      "epoch": 0.5298897666312049,
      "grad_norm": 0.5395227074623108,
      "learning_rate": 4.016081454833681e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 18845376,
      "step": 11705
    },
    {
      "epoch": 0.5301161185178479,
      "grad_norm": 0.3569139242172241,
      "learning_rate": 4.0153007160828245e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 18853088,
      "step": 11710
    },
    {
      "epoch": 0.5303424704044908,
      "grad_norm": 0.7251303195953369,
      "learning_rate": 4.0145197436535555e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 18861120,
      "step": 11715
    },
    {
      "epoch": 0.5305688222911338,
      "grad_norm": 1.2234644889831543,
      "learning_rate": 4.0137385376663095e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 18869696,
      "step": 11720
    },
    {
      "epoch": 0.5307951741777768,
      "grad_norm": 0.309627890586853,
      "learning_rate": 4.012957098241558e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 18877376,
      "step": 11725
    },
    {
      "epoch": 0.5310215260644198,
      "grad_norm": 0.36637434363365173,
      "learning_rate": 4.0121754254998076e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 18886080,
      "step": 11730
    },
    {
      "epoch": 0.5312478779510628,
      "grad_norm": 0.6117323040962219,
      "learning_rate": 4.011393519561606e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 18894304,
      "step": 11735
    },
    {
      "epoch": 0.5314742298377056,
      "grad_norm": 0.3875579535961151,
      "learning_rate": 4.010611380547529e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 18902528,
      "step": 11740
    },
    {
      "epoch": 0.5317005817243486,
      "grad_norm": 0.8271192312240601,
      "learning_rate": 4.009829008578192e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 18910304,
      "step": 11745
    },
    {
      "epoch": 0.5319269336109916,
      "grad_norm": 0.9495745897293091,
      "learning_rate": 4.00904640377425e-05,
      "loss": 0.1793,
      "num_input_tokens_seen": 18919072,
      "step": 11750
    },
    {
      "epoch": 0.5321532854976346,
      "grad_norm": 0.7616528868675232,
      "learning_rate": 4.0082635662563886e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 18926464,
      "step": 11755
    },
    {
      "epoch": 0.5323796373842776,
      "grad_norm": 0.4327675402164459,
      "learning_rate": 4.007480496145331e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 18935392,
      "step": 11760
    },
    {
      "epoch": 0.5326059892709206,
      "grad_norm": 0.6946629881858826,
      "learning_rate": 4.006697193561837e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 18943360,
      "step": 11765
    },
    {
      "epoch": 0.5328323411575635,
      "grad_norm": 1.2652758359909058,
      "learning_rate": 4.005913658626701e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 18951264,
      "step": 11770
    },
    {
      "epoch": 0.5330586930442065,
      "grad_norm": 0.9767701625823975,
      "learning_rate": 4.005129891460754e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 18959616,
      "step": 11775
    },
    {
      "epoch": 0.5332850449308495,
      "grad_norm": 0.3387017548084259,
      "learning_rate": 4.004345892184864e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 18967744,
      "step": 11780
    },
    {
      "epoch": 0.5335113968174925,
      "grad_norm": 0.3998875916004181,
      "learning_rate": 4.003561660919932e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 18975200,
      "step": 11785
    },
    {
      "epoch": 0.5337377487041355,
      "grad_norm": 0.47613146901130676,
      "learning_rate": 4.002777197786897e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 18982880,
      "step": 11790
    },
    {
      "epoch": 0.5339641005907785,
      "grad_norm": 0.7403627038002014,
      "learning_rate": 4.0019925029067326e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 18991008,
      "step": 11795
    },
    {
      "epoch": 0.5341904524774214,
      "grad_norm": 1.0982410907745361,
      "learning_rate": 4.0012075764004495e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 18999136,
      "step": 11800
    },
    {
      "epoch": 0.5341904524774214,
      "eval_loss": 0.15956808626651764,
      "eval_runtime": 404.9063,
      "eval_samples_per_second": 96.988,
      "eval_steps_per_second": 24.248,
      "num_input_tokens_seen": 18999136,
      "step": 11800
    },
    {
      "epoch": 0.5344168043640644,
      "grad_norm": 0.40682825446128845,
      "learning_rate": 4.000422418389094e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 19006752,
      "step": 11805
    },
    {
      "epoch": 0.5346431562507074,
      "grad_norm": 0.57232666015625,
      "learning_rate": 3.999637028993744e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 19014560,
      "step": 11810
    },
    {
      "epoch": 0.5348695081373503,
      "grad_norm": 0.3846348226070404,
      "learning_rate": 3.99885140833552e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 19022368,
      "step": 11815
    },
    {
      "epoch": 0.5350958600239933,
      "grad_norm": 0.6342281699180603,
      "learning_rate": 3.998065556535572e-05,
      "loss": 0.1114,
      "num_input_tokens_seen": 19030144,
      "step": 11820
    },
    {
      "epoch": 0.5353222119106362,
      "grad_norm": 0.40402689576148987,
      "learning_rate": 3.9972794737150895e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 19037760,
      "step": 11825
    },
    {
      "epoch": 0.5355485637972792,
      "grad_norm": 0.8865857124328613,
      "learning_rate": 3.996493159995297e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 19045440,
      "step": 11830
    },
    {
      "epoch": 0.5357749156839222,
      "grad_norm": 1.0493559837341309,
      "learning_rate": 3.995706615497453e-05,
      "loss": 0.1886,
      "num_input_tokens_seen": 19053344,
      "step": 11835
    },
    {
      "epoch": 0.5360012675705652,
      "grad_norm": 0.7039090991020203,
      "learning_rate": 3.994919840342852e-05,
      "loss": 0.1989,
      "num_input_tokens_seen": 19061216,
      "step": 11840
    },
    {
      "epoch": 0.5362276194572082,
      "grad_norm": 1.3529084920883179,
      "learning_rate": 3.994132834652825e-05,
      "loss": 0.2113,
      "num_input_tokens_seen": 19069024,
      "step": 11845
    },
    {
      "epoch": 0.5364539713438512,
      "grad_norm": 0.2969796657562256,
      "learning_rate": 3.99334559854874e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 19076832,
      "step": 11850
    },
    {
      "epoch": 0.5366803232304941,
      "grad_norm": 0.8147520422935486,
      "learning_rate": 3.9925581321519955e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 19084640,
      "step": 11855
    },
    {
      "epoch": 0.5369066751171371,
      "grad_norm": 0.9428049921989441,
      "learning_rate": 3.991770435584031e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 19092224,
      "step": 11860
    },
    {
      "epoch": 0.5371330270037801,
      "grad_norm": 0.8571634888648987,
      "learning_rate": 3.990982508966319e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 19100288,
      "step": 11865
    },
    {
      "epoch": 0.5373593788904231,
      "grad_norm": 0.4125654399394989,
      "learning_rate": 3.990194352420367e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 19107872,
      "step": 11870
    },
    {
      "epoch": 0.5375857307770661,
      "grad_norm": 0.4092762768268585,
      "learning_rate": 3.9894059660677184e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 19115680,
      "step": 11875
    },
    {
      "epoch": 0.5378120826637091,
      "grad_norm": 0.8224508762359619,
      "learning_rate": 3.9886173500299526e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 19123584,
      "step": 11880
    },
    {
      "epoch": 0.5380384345503519,
      "grad_norm": 1.1624151468276978,
      "learning_rate": 3.987828504428685e-05,
      "loss": 0.1041,
      "num_input_tokens_seen": 19132160,
      "step": 11885
    },
    {
      "epoch": 0.5382647864369949,
      "grad_norm": 0.4809432923793793,
      "learning_rate": 3.987039429385565e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 19140096,
      "step": 11890
    },
    {
      "epoch": 0.5384911383236379,
      "grad_norm": 0.7758264541625977,
      "learning_rate": 3.986250125022277e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 19148128,
      "step": 11895
    },
    {
      "epoch": 0.5387174902102809,
      "grad_norm": 0.9595478177070618,
      "learning_rate": 3.985460591460544e-05,
      "loss": 0.207,
      "num_input_tokens_seen": 19155872,
      "step": 11900
    },
    {
      "epoch": 0.5389438420969239,
      "grad_norm": 0.6056865453720093,
      "learning_rate": 3.984670828822118e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 19163936,
      "step": 11905
    },
    {
      "epoch": 0.5391701939835668,
      "grad_norm": 0.5014654397964478,
      "learning_rate": 3.983880837228794e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 19171616,
      "step": 11910
    },
    {
      "epoch": 0.5393965458702098,
      "grad_norm": 0.6099893450737,
      "learning_rate": 3.983090616802396e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 19180160,
      "step": 11915
    },
    {
      "epoch": 0.5396228977568528,
      "grad_norm": 0.37873396277427673,
      "learning_rate": 3.982300167664788e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 19188608,
      "step": 11920
    },
    {
      "epoch": 0.5398492496434958,
      "grad_norm": 0.4569169878959656,
      "learning_rate": 3.981509489937868e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 19196256,
      "step": 11925
    },
    {
      "epoch": 0.5400756015301388,
      "grad_norm": 1.0898746252059937,
      "learning_rate": 3.9807185837435643e-05,
      "loss": 0.0995,
      "num_input_tokens_seen": 19204256,
      "step": 11930
    },
    {
      "epoch": 0.5403019534167818,
      "grad_norm": 0.41616693139076233,
      "learning_rate": 3.9799274492038484e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 19212416,
      "step": 11935
    },
    {
      "epoch": 0.5405283053034247,
      "grad_norm": 0.503953754901886,
      "learning_rate": 3.979136086440722e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 19220160,
      "step": 11940
    },
    {
      "epoch": 0.5407546571900677,
      "grad_norm": 0.47670426964759827,
      "learning_rate": 3.9783444955762226e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 19228288,
      "step": 11945
    },
    {
      "epoch": 0.5409810090767107,
      "grad_norm": 1.0777231454849243,
      "learning_rate": 3.977552676732424e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 19235968,
      "step": 11950
    },
    {
      "epoch": 0.5412073609633536,
      "grad_norm": 0.8608806729316711,
      "learning_rate": 3.976760630031435e-05,
      "loss": 0.194,
      "num_input_tokens_seen": 19244384,
      "step": 11955
    },
    {
      "epoch": 0.5414337128499966,
      "grad_norm": 1.5645331144332886,
      "learning_rate": 3.975968355595398e-05,
      "loss": 0.2256,
      "num_input_tokens_seen": 19252192,
      "step": 11960
    },
    {
      "epoch": 0.5416600647366396,
      "grad_norm": 0.4947066307067871,
      "learning_rate": 3.9751758535464935e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 19259872,
      "step": 11965
    },
    {
      "epoch": 0.5418864166232825,
      "grad_norm": 0.44933006167411804,
      "learning_rate": 3.9743831240069326e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 19268032,
      "step": 11970
    },
    {
      "epoch": 0.5421127685099255,
      "grad_norm": 1.2526452541351318,
      "learning_rate": 3.9735901670989675e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 19276096,
      "step": 11975
    },
    {
      "epoch": 0.5423391203965685,
      "grad_norm": 0.5647858381271362,
      "learning_rate": 3.97279698294488e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 19284448,
      "step": 11980
    },
    {
      "epoch": 0.5425654722832115,
      "grad_norm": 0.6576390862464905,
      "learning_rate": 3.9720035716669876e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 19292096,
      "step": 11985
    },
    {
      "epoch": 0.5427918241698545,
      "grad_norm": 0.8363469839096069,
      "learning_rate": 3.9712099333876474e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 19300032,
      "step": 11990
    },
    {
      "epoch": 0.5430181760564974,
      "grad_norm": 0.8105944395065308,
      "learning_rate": 3.9704160682292475e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 19307968,
      "step": 11995
    },
    {
      "epoch": 0.5432445279431404,
      "grad_norm": 0.5183194875717163,
      "learning_rate": 3.9696219763142106e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 19316576,
      "step": 12000
    },
    {
      "epoch": 0.5432445279431404,
      "eval_loss": 0.15622811019420624,
      "eval_runtime": 404.3331,
      "eval_samples_per_second": 97.125,
      "eval_steps_per_second": 24.282,
      "num_input_tokens_seen": 19316576,
      "step": 12000
    },
    {
      "epoch": 0.5434708798297834,
      "grad_norm": 0.5024055242538452,
      "learning_rate": 3.968827657764997e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 19324320,
      "step": 12005
    },
    {
      "epoch": 0.5436972317164264,
      "grad_norm": 0.2904784381389618,
      "learning_rate": 3.9680331127041e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 19331616,
      "step": 12010
    },
    {
      "epoch": 0.5439235836030694,
      "grad_norm": 1.903767466545105,
      "learning_rate": 3.9672383412540495e-05,
      "loss": 0.2173,
      "num_input_tokens_seen": 19339840,
      "step": 12015
    },
    {
      "epoch": 0.5441499354897124,
      "grad_norm": 0.8110145926475525,
      "learning_rate": 3.966443343537407e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 19348064,
      "step": 12020
    },
    {
      "epoch": 0.5443762873763552,
      "grad_norm": 0.5759050846099854,
      "learning_rate": 3.965648119676772e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 19356192,
      "step": 12025
    },
    {
      "epoch": 0.5446026392629982,
      "grad_norm": 0.8469527363777161,
      "learning_rate": 3.96485266979478e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 19364576,
      "step": 12030
    },
    {
      "epoch": 0.5448289911496412,
      "grad_norm": 0.7690746188163757,
      "learning_rate": 3.9640569940140974e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 19372640,
      "step": 12035
    },
    {
      "epoch": 0.5450553430362842,
      "grad_norm": 0.38469868898391724,
      "learning_rate": 3.963261092457428e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 19380736,
      "step": 12040
    },
    {
      "epoch": 0.5452816949229272,
      "grad_norm": 0.6604794263839722,
      "learning_rate": 3.962464965247509e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 19389120,
      "step": 12045
    },
    {
      "epoch": 0.5455080468095702,
      "grad_norm": 1.0864156484603882,
      "learning_rate": 3.9616686125071135e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 19396832,
      "step": 12050
    },
    {
      "epoch": 0.5457343986962131,
      "grad_norm": 0.653961718082428,
      "learning_rate": 3.9608720343590506e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 19404576,
      "step": 12055
    },
    {
      "epoch": 0.5459607505828561,
      "grad_norm": 1.016191840171814,
      "learning_rate": 3.960075230926161e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 19412480,
      "step": 12060
    },
    {
      "epoch": 0.5461871024694991,
      "grad_norm": 0.5968104600906372,
      "learning_rate": 3.959278202331322e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 19420960,
      "step": 12065
    },
    {
      "epoch": 0.5464134543561421,
      "grad_norm": 0.7075677514076233,
      "learning_rate": 3.958480948697446e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 19429376,
      "step": 12070
    },
    {
      "epoch": 0.5466398062427851,
      "grad_norm": 0.5626567006111145,
      "learning_rate": 3.95768347014748e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 19437600,
      "step": 12075
    },
    {
      "epoch": 0.546866158129428,
      "grad_norm": 0.7187651991844177,
      "learning_rate": 3.956885766804404e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 19445600,
      "step": 12080
    },
    {
      "epoch": 0.547092510016071,
      "grad_norm": 0.9685410857200623,
      "learning_rate": 3.956087838791235e-05,
      "loss": 0.2134,
      "num_input_tokens_seen": 19453664,
      "step": 12085
    },
    {
      "epoch": 0.547318861902714,
      "grad_norm": 0.9460963010787964,
      "learning_rate": 3.955289686231022e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 19462208,
      "step": 12090
    },
    {
      "epoch": 0.5475452137893569,
      "grad_norm": 0.9830073118209839,
      "learning_rate": 3.9544913092468504e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 19470176,
      "step": 12095
    },
    {
      "epoch": 0.5477715656759999,
      "grad_norm": 0.7391504645347595,
      "learning_rate": 3.9536927079618425e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 19478048,
      "step": 12100
    },
    {
      "epoch": 0.5479979175626429,
      "grad_norm": 0.8764104843139648,
      "learning_rate": 3.9528938824991494e-05,
      "loss": 0.1878,
      "num_input_tokens_seen": 19485728,
      "step": 12105
    },
    {
      "epoch": 0.5482242694492858,
      "grad_norm": 0.6011919379234314,
      "learning_rate": 3.952094832981962e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 19493216,
      "step": 12110
    },
    {
      "epoch": 0.5484506213359288,
      "grad_norm": 0.5332769155502319,
      "learning_rate": 3.951295559533503e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 19501280,
      "step": 12115
    },
    {
      "epoch": 0.5486769732225718,
      "grad_norm": 0.44070667028427124,
      "learning_rate": 3.95049606227703e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 19509056,
      "step": 12120
    },
    {
      "epoch": 0.5489033251092148,
      "grad_norm": 0.7304631471633911,
      "learning_rate": 3.949696341335838e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 19517344,
      "step": 12125
    },
    {
      "epoch": 0.5491296769958578,
      "grad_norm": 0.4046960473060608,
      "learning_rate": 3.9488963968332503e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 19525792,
      "step": 12130
    },
    {
      "epoch": 0.5493560288825008,
      "grad_norm": 0.8096943497657776,
      "learning_rate": 3.948096228892631e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 19533792,
      "step": 12135
    },
    {
      "epoch": 0.5495823807691437,
      "grad_norm": 0.4013364315032959,
      "learning_rate": 3.947295837637375e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 19541888,
      "step": 12140
    },
    {
      "epoch": 0.5498087326557867,
      "grad_norm": 1.4597989320755005,
      "learning_rate": 3.9464952231909135e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 19550784,
      "step": 12145
    },
    {
      "epoch": 0.5500350845424297,
      "grad_norm": 0.5005800127983093,
      "learning_rate": 3.945694385676711e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 19558272,
      "step": 12150
    },
    {
      "epoch": 0.5502614364290727,
      "grad_norm": 0.8360502123832703,
      "learning_rate": 3.944893325218265e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 19565984,
      "step": 12155
    },
    {
      "epoch": 0.5504877883157157,
      "grad_norm": 0.8564375638961792,
      "learning_rate": 3.944092041939112e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 19574880,
      "step": 12160
    },
    {
      "epoch": 0.5507141402023585,
      "grad_norm": 0.8940531611442566,
      "learning_rate": 3.943290535962818e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 19583584,
      "step": 12165
    },
    {
      "epoch": 0.5509404920890015,
      "grad_norm": 0.37481093406677246,
      "learning_rate": 3.942488807412985e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 19591552,
      "step": 12170
    },
    {
      "epoch": 0.5511668439756445,
      "grad_norm": 0.49159762263298035,
      "learning_rate": 3.941686856413251e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 19599168,
      "step": 12175
    },
    {
      "epoch": 0.5513931958622875,
      "grad_norm": 0.883547306060791,
      "learning_rate": 3.9408846830872874e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 19606912,
      "step": 12180
    },
    {
      "epoch": 0.5516195477489305,
      "grad_norm": 0.5254772305488586,
      "learning_rate": 3.940082287558798e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 19615168,
      "step": 12185
    },
    {
      "epoch": 0.5518458996355735,
      "grad_norm": 0.650134265422821,
      "learning_rate": 3.939279669951522e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 19623744,
      "step": 12190
    },
    {
      "epoch": 0.5520722515222164,
      "grad_norm": 0.33024728298187256,
      "learning_rate": 3.938476830389234e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 19632832,
      "step": 12195
    },
    {
      "epoch": 0.5522986034088594,
      "grad_norm": 0.915336012840271,
      "learning_rate": 3.937673768995742e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 19640544,
      "step": 12200
    },
    {
      "epoch": 0.5522986034088594,
      "eval_loss": 0.15512973070144653,
      "eval_runtime": 405.1632,
      "eval_samples_per_second": 96.926,
      "eval_steps_per_second": 24.232,
      "num_input_tokens_seen": 19640544,
      "step": 12200
    },
    {
      "epoch": 0.5525249552955024,
      "grad_norm": 0.7659688591957092,
      "learning_rate": 3.936870485894888e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 19648704,
      "step": 12205
    },
    {
      "epoch": 0.5527513071821454,
      "grad_norm": 0.7345837950706482,
      "learning_rate": 3.9360669812105475e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 19656416,
      "step": 12210
    },
    {
      "epoch": 0.5529776590687884,
      "grad_norm": 0.4557288885116577,
      "learning_rate": 3.9352632550666325e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 19664672,
      "step": 12215
    },
    {
      "epoch": 0.5532040109554314,
      "grad_norm": 0.45783087611198425,
      "learning_rate": 3.9344593075870866e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 19672704,
      "step": 12220
    },
    {
      "epoch": 0.5534303628420743,
      "grad_norm": 0.31112539768218994,
      "learning_rate": 3.933655138895889e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 19680960,
      "step": 12225
    },
    {
      "epoch": 0.5536567147287172,
      "grad_norm": 0.7508276700973511,
      "learning_rate": 3.932850749117053e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 19689344,
      "step": 12230
    },
    {
      "epoch": 0.5538830666153602,
      "grad_norm": 0.7853493690490723,
      "learning_rate": 3.932046138374624e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 19697312,
      "step": 12235
    },
    {
      "epoch": 0.5541094185020032,
      "grad_norm": 1.1749439239501953,
      "learning_rate": 3.9312413067926854e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 19705056,
      "step": 12240
    },
    {
      "epoch": 0.5543357703886462,
      "grad_norm": 0.48492860794067383,
      "learning_rate": 3.9304362544953506e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 19713024,
      "step": 12245
    },
    {
      "epoch": 0.5545621222752891,
      "grad_norm": 0.812837541103363,
      "learning_rate": 3.929630981606769e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 19720800,
      "step": 12250
    },
    {
      "epoch": 0.5547884741619321,
      "grad_norm": 0.5547254681587219,
      "learning_rate": 3.928825488251124e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 19728672,
      "step": 12255
    },
    {
      "epoch": 0.5550148260485751,
      "grad_norm": 0.47715023159980774,
      "learning_rate": 3.9280197745526344e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 19736544,
      "step": 12260
    },
    {
      "epoch": 0.5552411779352181,
      "grad_norm": 0.5995262265205383,
      "learning_rate": 3.9272138406355495e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 19745312,
      "step": 12265
    },
    {
      "epoch": 0.5554675298218611,
      "grad_norm": 1.281214714050293,
      "learning_rate": 3.926407686624154e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 19753248,
      "step": 12270
    },
    {
      "epoch": 0.5556938817085041,
      "grad_norm": 0.5527825355529785,
      "learning_rate": 3.9256013126427684e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 19761280,
      "step": 12275
    },
    {
      "epoch": 0.555920233595147,
      "grad_norm": 0.5273796916007996,
      "learning_rate": 3.9247947188157455e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 19769376,
      "step": 12280
    },
    {
      "epoch": 0.55614658548179,
      "grad_norm": 0.5058216452598572,
      "learning_rate": 3.9239879052674715e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 19777600,
      "step": 12285
    },
    {
      "epoch": 0.556372937368433,
      "grad_norm": 1.5408648252487183,
      "learning_rate": 3.9231808721223673e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 19785760,
      "step": 12290
    },
    {
      "epoch": 0.556599289255076,
      "grad_norm": 0.5364517569541931,
      "learning_rate": 3.9223736195048886e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 19793472,
      "step": 12295
    },
    {
      "epoch": 0.556825641141719,
      "grad_norm": 0.5107898712158203,
      "learning_rate": 3.921566147539523e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 19801376,
      "step": 12300
    },
    {
      "epoch": 0.5570519930283618,
      "grad_norm": 0.8361883163452148,
      "learning_rate": 3.920758456350792e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 19809088,
      "step": 12305
    },
    {
      "epoch": 0.5572783449150048,
      "grad_norm": 0.7294726967811584,
      "learning_rate": 3.919950546063253e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 19816960,
      "step": 12310
    },
    {
      "epoch": 0.5575046968016478,
      "grad_norm": 0.5562055110931396,
      "learning_rate": 3.919142416801496e-05,
      "loss": 0.2019,
      "num_input_tokens_seen": 19825152,
      "step": 12315
    },
    {
      "epoch": 0.5577310486882908,
      "grad_norm": 0.7164984345436096,
      "learning_rate": 3.918334068690144e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 19833280,
      "step": 12320
    },
    {
      "epoch": 0.5579574005749338,
      "grad_norm": 1.2792892456054688,
      "learning_rate": 3.917525501853855e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 19841312,
      "step": 12325
    },
    {
      "epoch": 0.5581837524615768,
      "grad_norm": 0.7934985756874084,
      "learning_rate": 3.916716716417319e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 19848864,
      "step": 12330
    },
    {
      "epoch": 0.5584101043482197,
      "grad_norm": 0.9713300466537476,
      "learning_rate": 3.915907712505263e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 19856864,
      "step": 12335
    },
    {
      "epoch": 0.5586364562348627,
      "grad_norm": 0.5705817341804504,
      "learning_rate": 3.915098490242444e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 19864608,
      "step": 12340
    },
    {
      "epoch": 0.5588628081215057,
      "grad_norm": 0.869909942150116,
      "learning_rate": 3.914289049753654e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 19872480,
      "step": 12345
    },
    {
      "epoch": 0.5590891600081487,
      "grad_norm": 0.3824942111968994,
      "learning_rate": 3.913479391163719e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 19880480,
      "step": 12350
    },
    {
      "epoch": 0.5593155118947917,
      "grad_norm": 1.4056719541549683,
      "learning_rate": 3.9126695145975e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 19888576,
      "step": 12355
    },
    {
      "epoch": 0.5595418637814347,
      "grad_norm": 0.28677311539649963,
      "learning_rate": 3.911859420179889e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 19896416,
      "step": 12360
    },
    {
      "epoch": 0.5597682156680776,
      "grad_norm": 0.8500235080718994,
      "learning_rate": 3.911049108035813e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 19904160,
      "step": 12365
    },
    {
      "epoch": 0.5599945675547205,
      "grad_norm": 0.3780805468559265,
      "learning_rate": 3.910238578290232e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 19911936,
      "step": 12370
    },
    {
      "epoch": 0.5602209194413635,
      "grad_norm": 0.5934856534004211,
      "learning_rate": 3.90942783106814e-05,
      "loss": 0.1553,
      "num_input_tokens_seen": 19920128,
      "step": 12375
    },
    {
      "epoch": 0.5604472713280065,
      "grad_norm": 1.0372742414474487,
      "learning_rate": 3.908616866494564e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 19928320,
      "step": 12380
    },
    {
      "epoch": 0.5606736232146495,
      "grad_norm": 0.4824271500110626,
      "learning_rate": 3.907805684694566e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 19935616,
      "step": 12385
    },
    {
      "epoch": 0.5608999751012924,
      "grad_norm": 1.1938878297805786,
      "learning_rate": 3.90699428579324e-05,
      "loss": 0.192,
      "num_input_tokens_seen": 19943360,
      "step": 12390
    },
    {
      "epoch": 0.5611263269879354,
      "grad_norm": 0.7359627485275269,
      "learning_rate": 3.906182669915713e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 19951712,
      "step": 12395
    },
    {
      "epoch": 0.5613526788745784,
      "grad_norm": 0.6926642060279846,
      "learning_rate": 3.9053708371871476e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 19959648,
      "step": 12400
    },
    {
      "epoch": 0.5613526788745784,
      "eval_loss": 0.15454775094985962,
      "eval_runtime": 404.7249,
      "eval_samples_per_second": 97.031,
      "eval_steps_per_second": 24.258,
      "num_input_tokens_seen": 19959648,
      "step": 12400
    },
    {
      "epoch": 0.5615790307612214,
      "grad_norm": 0.7573134899139404,
      "learning_rate": 3.904558787732738e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 19967840,
      "step": 12405
    },
    {
      "epoch": 0.5618053826478644,
      "grad_norm": 0.6584493517875671,
      "learning_rate": 3.9037465216777135e-05,
      "loss": 0.1857,
      "num_input_tokens_seen": 19975648,
      "step": 12410
    },
    {
      "epoch": 0.5620317345345074,
      "grad_norm": 0.5978819727897644,
      "learning_rate": 3.902934039147334e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 19983584,
      "step": 12415
    },
    {
      "epoch": 0.5622580864211503,
      "grad_norm": 0.6355190277099609,
      "learning_rate": 3.902121340266894e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 19991744,
      "step": 12420
    },
    {
      "epoch": 0.5624844383077933,
      "grad_norm": 0.47284677624702454,
      "learning_rate": 3.9013084251617246e-05,
      "loss": 0.2233,
      "num_input_tokens_seen": 20000352,
      "step": 12425
    },
    {
      "epoch": 0.5627107901944363,
      "grad_norm": 0.6408313512802124,
      "learning_rate": 3.9004952939571865e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 20008192,
      "step": 12430
    },
    {
      "epoch": 0.5629371420810793,
      "grad_norm": 0.6008126735687256,
      "learning_rate": 3.899681946778673e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 20016576,
      "step": 12435
    },
    {
      "epoch": 0.5631634939677223,
      "grad_norm": 0.805709183216095,
      "learning_rate": 3.898868383751615e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 20024576,
      "step": 12440
    },
    {
      "epoch": 0.5633898458543652,
      "grad_norm": 0.410254567861557,
      "learning_rate": 3.8980546050014724e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 20032352,
      "step": 12445
    },
    {
      "epoch": 0.5636161977410081,
      "grad_norm": 0.4921114444732666,
      "learning_rate": 3.897240610653741e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 20040416,
      "step": 12450
    },
    {
      "epoch": 0.5638425496276511,
      "grad_norm": 0.6273542642593384,
      "learning_rate": 3.896426400833948e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 20047936,
      "step": 12455
    },
    {
      "epoch": 0.5640689015142941,
      "grad_norm": 0.7380044460296631,
      "learning_rate": 3.895611975667656e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 20055520,
      "step": 12460
    },
    {
      "epoch": 0.5642952534009371,
      "grad_norm": 0.7099901437759399,
      "learning_rate": 3.8947973352804584e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 20063008,
      "step": 12465
    },
    {
      "epoch": 0.5645216052875801,
      "grad_norm": 0.8433166146278381,
      "learning_rate": 3.893982479797984e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 20071040,
      "step": 12470
    },
    {
      "epoch": 0.564747957174223,
      "grad_norm": 1.0068469047546387,
      "learning_rate": 3.8931674093458926e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 20078464,
      "step": 12475
    },
    {
      "epoch": 0.564974309060866,
      "grad_norm": 0.5681838989257812,
      "learning_rate": 3.89235212404988e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 20086400,
      "step": 12480
    },
    {
      "epoch": 0.565200660947509,
      "grad_norm": 1.4816313982009888,
      "learning_rate": 3.891536624035672e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 20094592,
      "step": 12485
    },
    {
      "epoch": 0.565427012834152,
      "grad_norm": 0.6090088486671448,
      "learning_rate": 3.8907209094290295e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 20102880,
      "step": 12490
    },
    {
      "epoch": 0.565653364720795,
      "grad_norm": 0.6116510629653931,
      "learning_rate": 3.8899049803557466e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 20111872,
      "step": 12495
    },
    {
      "epoch": 0.565879716607438,
      "grad_norm": 0.567430317401886,
      "learning_rate": 3.889088836941648e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 20119648,
      "step": 12500
    },
    {
      "epoch": 0.5661060684940808,
      "grad_norm": 0.4141646921634674,
      "learning_rate": 3.8882724793125946e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 20128096,
      "step": 12505
    },
    {
      "epoch": 0.5663324203807238,
      "grad_norm": 0.48030561208724976,
      "learning_rate": 3.8874559075944794e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 20136224,
      "step": 12510
    },
    {
      "epoch": 0.5665587722673668,
      "grad_norm": 0.5591956377029419,
      "learning_rate": 3.886639121913227e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 20144000,
      "step": 12515
    },
    {
      "epoch": 0.5667851241540098,
      "grad_norm": 0.7877773642539978,
      "learning_rate": 3.885822122394797e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 20151328,
      "step": 12520
    },
    {
      "epoch": 0.5670114760406528,
      "grad_norm": 0.4767967462539673,
      "learning_rate": 3.8850049091651794e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 20159456,
      "step": 12525
    },
    {
      "epoch": 0.5672378279272958,
      "grad_norm": 0.8555184602737427,
      "learning_rate": 3.8841874823504e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 20168032,
      "step": 12530
    },
    {
      "epoch": 0.5674641798139387,
      "grad_norm": 0.4786396026611328,
      "learning_rate": 3.8833698420765157e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 20176256,
      "step": 12535
    },
    {
      "epoch": 0.5676905317005817,
      "grad_norm": 0.3516573905944824,
      "learning_rate": 3.882551988469618e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 20184416,
      "step": 12540
    },
    {
      "epoch": 0.5679168835872247,
      "grad_norm": 0.7383844256401062,
      "learning_rate": 3.881733921655829e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 20192416,
      "step": 12545
    },
    {
      "epoch": 0.5681432354738677,
      "grad_norm": 0.8876021504402161,
      "learning_rate": 3.8809156417613054e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 20200512,
      "step": 12550
    },
    {
      "epoch": 0.5683695873605107,
      "grad_norm": 0.6538239121437073,
      "learning_rate": 3.8800971489122364e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 20208544,
      "step": 12555
    },
    {
      "epoch": 0.5685959392471536,
      "grad_norm": 1.0264928340911865,
      "learning_rate": 3.8792784432348434e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 20216224,
      "step": 12560
    },
    {
      "epoch": 0.5688222911337966,
      "grad_norm": 0.27005183696746826,
      "learning_rate": 3.878459524855381e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 20223520,
      "step": 12565
    },
    {
      "epoch": 0.5690486430204396,
      "grad_norm": 0.5440042018890381,
      "learning_rate": 3.8776403939001384e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 20231616,
      "step": 12570
    },
    {
      "epoch": 0.5692749949070826,
      "grad_norm": 0.791841447353363,
      "learning_rate": 3.876821050495433e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 20239296,
      "step": 12575
    },
    {
      "epoch": 0.5695013467937255,
      "grad_norm": 0.40781810879707336,
      "learning_rate": 3.87600149476762e-05,
      "loss": 0.1199,
      "num_input_tokens_seen": 20246976,
      "step": 12580
    },
    {
      "epoch": 0.5697276986803685,
      "grad_norm": 0.43031224608421326,
      "learning_rate": 3.8751817268430843e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 20254784,
      "step": 12585
    },
    {
      "epoch": 0.5699540505670114,
      "grad_norm": 0.4243165850639343,
      "learning_rate": 3.8743617468482464e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 20262656,
      "step": 12590
    },
    {
      "epoch": 0.5701804024536544,
      "grad_norm": 0.5087380409240723,
      "learning_rate": 3.8735415549095535e-05,
      "loss": 0.2026,
      "num_input_tokens_seen": 20270688,
      "step": 12595
    },
    {
      "epoch": 0.5704067543402974,
      "grad_norm": 0.8054437637329102,
      "learning_rate": 3.8727211511534934e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 20279232,
      "step": 12600
    },
    {
      "epoch": 0.5704067543402974,
      "eval_loss": 0.15435658395290375,
      "eval_runtime": 404.5629,
      "eval_samples_per_second": 97.07,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 20279232,
      "step": 12600
    },
    {
      "epoch": 0.5706331062269404,
      "grad_norm": 0.5244632959365845,
      "learning_rate": 3.8719005357065804e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 20287584,
      "step": 12605
    },
    {
      "epoch": 0.5708594581135834,
      "grad_norm": 0.3656691014766693,
      "learning_rate": 3.8710797086953645e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 20295328,
      "step": 12610
    },
    {
      "epoch": 0.5710858100002264,
      "grad_norm": 0.5939803719520569,
      "learning_rate": 3.870258670246427e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 20303328,
      "step": 12615
    },
    {
      "epoch": 0.5713121618868693,
      "grad_norm": 0.45448800921440125,
      "learning_rate": 3.869437420486384e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 20311392,
      "step": 12620
    },
    {
      "epoch": 0.5715385137735123,
      "grad_norm": 0.5846982002258301,
      "learning_rate": 3.8686159595418805e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 20320224,
      "step": 12625
    },
    {
      "epoch": 0.5717648656601553,
      "grad_norm": 0.6702300906181335,
      "learning_rate": 3.867794287539597e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 20328128,
      "step": 12630
    },
    {
      "epoch": 0.5719912175467983,
      "grad_norm": 0.5940059423446655,
      "learning_rate": 3.866972404606245e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 20336000,
      "step": 12635
    },
    {
      "epoch": 0.5722175694334413,
      "grad_norm": 1.1698250770568848,
      "learning_rate": 3.866150310868571e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 20344192,
      "step": 12640
    },
    {
      "epoch": 0.5724439213200841,
      "grad_norm": 0.5007466673851013,
      "learning_rate": 3.8653280064533506e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 20351872,
      "step": 12645
    },
    {
      "epoch": 0.5726702732067271,
      "grad_norm": 0.5765907168388367,
      "learning_rate": 3.864505491487394e-05,
      "loss": 0.1903,
      "num_input_tokens_seen": 20360384,
      "step": 12650
    },
    {
      "epoch": 0.5728966250933701,
      "grad_norm": 1.5431458950042725,
      "learning_rate": 3.8636827660975414e-05,
      "loss": 0.1858,
      "num_input_tokens_seen": 20369120,
      "step": 12655
    },
    {
      "epoch": 0.5731229769800131,
      "grad_norm": 1.0068498849868774,
      "learning_rate": 3.862859830410671e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 20377568,
      "step": 12660
    },
    {
      "epoch": 0.5733493288666561,
      "grad_norm": 0.851453959941864,
      "learning_rate": 3.862036684553688e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 20385984,
      "step": 12665
    },
    {
      "epoch": 0.5735756807532991,
      "grad_norm": 0.8289304971694946,
      "learning_rate": 3.8612133286535314e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 20394016,
      "step": 12670
    },
    {
      "epoch": 0.573802032639942,
      "grad_norm": 0.44921186566352844,
      "learning_rate": 3.860389762837173e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 20402304,
      "step": 12675
    },
    {
      "epoch": 0.574028384526585,
      "grad_norm": 0.7036224603652954,
      "learning_rate": 3.859565987231618e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 20410528,
      "step": 12680
    },
    {
      "epoch": 0.574254736413228,
      "grad_norm": 0.9409956932067871,
      "learning_rate": 3.858742001963902e-05,
      "loss": 0.1762,
      "num_input_tokens_seen": 20418432,
      "step": 12685
    },
    {
      "epoch": 0.574481088299871,
      "grad_norm": 0.5376362800598145,
      "learning_rate": 3.857917807161094e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 20425856,
      "step": 12690
    },
    {
      "epoch": 0.574707440186514,
      "grad_norm": 0.4670591354370117,
      "learning_rate": 3.857093402950296e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 20433600,
      "step": 12695
    },
    {
      "epoch": 0.574933792073157,
      "grad_norm": 0.5757614970207214,
      "learning_rate": 3.8562687894586414e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 20441056,
      "step": 12700
    },
    {
      "epoch": 0.5751601439597999,
      "grad_norm": 0.759824275970459,
      "learning_rate": 3.8554439668132946e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 20448704,
      "step": 12705
    },
    {
      "epoch": 0.5753864958464429,
      "grad_norm": 1.2345783710479736,
      "learning_rate": 3.854618935141455e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 20456768,
      "step": 12710
    },
    {
      "epoch": 0.5756128477330859,
      "grad_norm": 0.5176426768302917,
      "learning_rate": 3.8537936945703525e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 20464320,
      "step": 12715
    },
    {
      "epoch": 0.5758391996197288,
      "grad_norm": 0.3873089551925659,
      "learning_rate": 3.852968245227249e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 20472416,
      "step": 12720
    },
    {
      "epoch": 0.5760655515063718,
      "grad_norm": 0.5508853197097778,
      "learning_rate": 3.85214258723944e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 20480896,
      "step": 12725
    },
    {
      "epoch": 0.5762919033930147,
      "grad_norm": 1.1417142152786255,
      "learning_rate": 3.8513167207342524e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 20489632,
      "step": 12730
    },
    {
      "epoch": 0.5765182552796577,
      "grad_norm": 0.33418041467666626,
      "learning_rate": 3.850490645839044e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 20497376,
      "step": 12735
    },
    {
      "epoch": 0.5767446071663007,
      "grad_norm": 0.4879609942436218,
      "learning_rate": 3.849664362681207e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 20505184,
      "step": 12740
    },
    {
      "epoch": 0.5769709590529437,
      "grad_norm": 0.8465380072593689,
      "learning_rate": 3.848837871388165e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 20513568,
      "step": 12745
    },
    {
      "epoch": 0.5771973109395867,
      "grad_norm": 0.7495355010032654,
      "learning_rate": 3.848011172087371e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 20522016,
      "step": 12750
    },
    {
      "epoch": 0.5774236628262297,
      "grad_norm": 0.5122522711753845,
      "learning_rate": 3.847184264906315e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 20529696,
      "step": 12755
    },
    {
      "epoch": 0.5776500147128726,
      "grad_norm": 1.0752533674240112,
      "learning_rate": 3.846357149972516e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 20537248,
      "step": 12760
    },
    {
      "epoch": 0.5778763665995156,
      "grad_norm": 0.6287433505058289,
      "learning_rate": 3.8455298274135246e-05,
      "loss": 0.1782,
      "num_input_tokens_seen": 20544960,
      "step": 12765
    },
    {
      "epoch": 0.5781027184861586,
      "grad_norm": 0.8054229021072388,
      "learning_rate": 3.8447022973569254e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 20552704,
      "step": 12770
    },
    {
      "epoch": 0.5783290703728016,
      "grad_norm": 0.58626389503479,
      "learning_rate": 3.843874559930332e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 20561120,
      "step": 12775
    },
    {
      "epoch": 0.5785554222594446,
      "grad_norm": 0.8787646889686584,
      "learning_rate": 3.843046615261394e-05,
      "loss": 0.1298,
      "num_input_tokens_seen": 20569312,
      "step": 12780
    },
    {
      "epoch": 0.5787817741460876,
      "grad_norm": 0.5066097378730774,
      "learning_rate": 3.842218463477791e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 20576992,
      "step": 12785
    },
    {
      "epoch": 0.5790081260327304,
      "grad_norm": 0.4376106858253479,
      "learning_rate": 3.841390104707233e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 20584896,
      "step": 12790
    },
    {
      "epoch": 0.5792344779193734,
      "grad_norm": 0.6245453953742981,
      "learning_rate": 3.8405615390774643e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 20593792,
      "step": 12795
    },
    {
      "epoch": 0.5794608298060164,
      "grad_norm": 0.8310851454734802,
      "learning_rate": 3.839732766716259e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 20601792,
      "step": 12800
    },
    {
      "epoch": 0.5794608298060164,
      "eval_loss": 0.1563018411397934,
      "eval_runtime": 404.4417,
      "eval_samples_per_second": 97.099,
      "eval_steps_per_second": 24.275,
      "num_input_tokens_seen": 20601792,
      "step": 12800
    },
    {
      "epoch": 0.5796871816926594,
      "grad_norm": 0.4754863679409027,
      "learning_rate": 3.838903787751425e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 20610048,
      "step": 12805
    },
    {
      "epoch": 0.5799135335793024,
      "grad_norm": 0.7662886381149292,
      "learning_rate": 3.838074602310802e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 20618240,
      "step": 12810
    },
    {
      "epoch": 0.5801398854659453,
      "grad_norm": 0.5950760841369629,
      "learning_rate": 3.837245210522258e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 20626240,
      "step": 12815
    },
    {
      "epoch": 0.5803662373525883,
      "grad_norm": 0.7595810294151306,
      "learning_rate": 3.8364156125136996e-05,
      "loss": 0.1709,
      "num_input_tokens_seen": 20634400,
      "step": 12820
    },
    {
      "epoch": 0.5805925892392313,
      "grad_norm": 0.31097733974456787,
      "learning_rate": 3.835585808413059e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 20642272,
      "step": 12825
    },
    {
      "epoch": 0.5808189411258743,
      "grad_norm": 0.5514848828315735,
      "learning_rate": 3.8347557983483024e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 20649952,
      "step": 12830
    },
    {
      "epoch": 0.5810452930125173,
      "grad_norm": 1.040112853050232,
      "learning_rate": 3.833925582447428e-05,
      "loss": 0.1785,
      "num_input_tokens_seen": 20657760,
      "step": 12835
    },
    {
      "epoch": 0.5812716448991603,
      "grad_norm": 0.4919884204864502,
      "learning_rate": 3.8330951608384656e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 20666016,
      "step": 12840
    },
    {
      "epoch": 0.5814979967858032,
      "grad_norm": 0.8773617148399353,
      "learning_rate": 3.832264533649477e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 20674240,
      "step": 12845
    },
    {
      "epoch": 0.5817243486724462,
      "grad_norm": 0.41869670152664185,
      "learning_rate": 3.8314337010085555e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 20682944,
      "step": 12850
    },
    {
      "epoch": 0.5819507005590892,
      "grad_norm": 0.5739194750785828,
      "learning_rate": 3.830602663043824e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 20690976,
      "step": 12855
    },
    {
      "epoch": 0.5821770524457321,
      "grad_norm": 0.7233884930610657,
      "learning_rate": 3.8297714198834414e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 20698880,
      "step": 12860
    },
    {
      "epoch": 0.5824034043323751,
      "grad_norm": 0.5603764653205872,
      "learning_rate": 3.828939971655595e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 20706720,
      "step": 12865
    },
    {
      "epoch": 0.5826297562190181,
      "grad_norm": 0.832326352596283,
      "learning_rate": 3.828108318488505e-05,
      "loss": 0.2103,
      "num_input_tokens_seen": 20714688,
      "step": 12870
    },
    {
      "epoch": 0.582856108105661,
      "grad_norm": 0.4653598666191101,
      "learning_rate": 3.8272764605104216e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 20722336,
      "step": 12875
    },
    {
      "epoch": 0.583082459992304,
      "grad_norm": 1.5493295192718506,
      "learning_rate": 3.826444397849628e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 20730560,
      "step": 12880
    },
    {
      "epoch": 0.583308811878947,
      "grad_norm": 0.43813297152519226,
      "learning_rate": 3.825612130634439e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 20738688,
      "step": 12885
    },
    {
      "epoch": 0.58353516376559,
      "grad_norm": 0.4706985354423523,
      "learning_rate": 3.824779658993202e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 20746496,
      "step": 12890
    },
    {
      "epoch": 0.583761515652233,
      "grad_norm": 0.5703331232070923,
      "learning_rate": 3.823946983054292e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 20754656,
      "step": 12895
    },
    {
      "epoch": 0.5839878675388759,
      "grad_norm": 0.8144691586494446,
      "learning_rate": 3.82311410294612e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 20762432,
      "step": 12900
    },
    {
      "epoch": 0.5842142194255189,
      "grad_norm": 0.47464385628700256,
      "learning_rate": 3.822281018797127e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 20770336,
      "step": 12905
    },
    {
      "epoch": 0.5844405713121619,
      "grad_norm": 1.3110519647598267,
      "learning_rate": 3.821447730735783e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 20778368,
      "step": 12910
    },
    {
      "epoch": 0.5846669231988049,
      "grad_norm": 0.9618526697158813,
      "learning_rate": 3.820614238890592e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 20786752,
      "step": 12915
    },
    {
      "epoch": 0.5848932750854479,
      "grad_norm": 0.6942980289459229,
      "learning_rate": 3.819780543390091e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 20794400,
      "step": 12920
    },
    {
      "epoch": 0.5851196269720909,
      "grad_norm": 0.47580409049987793,
      "learning_rate": 3.818946644362844e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 20802656,
      "step": 12925
    },
    {
      "epoch": 0.5853459788587337,
      "grad_norm": 0.36273500323295593,
      "learning_rate": 3.81811254193745e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 20810400,
      "step": 12930
    },
    {
      "epoch": 0.5855723307453767,
      "grad_norm": 0.6240756511688232,
      "learning_rate": 3.8172782362425366e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 20818144,
      "step": 12935
    },
    {
      "epoch": 0.5857986826320197,
      "grad_norm": 0.37807488441467285,
      "learning_rate": 3.816443727406765e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 20826496,
      "step": 12940
    },
    {
      "epoch": 0.5860250345186627,
      "grad_norm": 0.9399060010910034,
      "learning_rate": 3.815609015558829e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 20835136,
      "step": 12945
    },
    {
      "epoch": 0.5862513864053057,
      "grad_norm": 1.0935513973236084,
      "learning_rate": 3.814774100827448e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 20843072,
      "step": 12950
    },
    {
      "epoch": 0.5864777382919487,
      "grad_norm": 0.5408492088317871,
      "learning_rate": 3.813938983341379e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 20851232,
      "step": 12955
    },
    {
      "epoch": 0.5867040901785916,
      "grad_norm": 0.725620687007904,
      "learning_rate": 3.813103663229407e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 20859232,
      "step": 12960
    },
    {
      "epoch": 0.5869304420652346,
      "grad_norm": 0.9979696273803711,
      "learning_rate": 3.812268140620349e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 20867456,
      "step": 12965
    },
    {
      "epoch": 0.5871567939518776,
      "grad_norm": 1.3121713399887085,
      "learning_rate": 3.811432415643051e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 20875296,
      "step": 12970
    },
    {
      "epoch": 0.5873831458385206,
      "grad_norm": 0.7476275563240051,
      "learning_rate": 3.8105964884263954e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 20883168,
      "step": 12975
    },
    {
      "epoch": 0.5876094977251636,
      "grad_norm": 0.5852876901626587,
      "learning_rate": 3.809760359099291e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 20891168,
      "step": 12980
    },
    {
      "epoch": 0.5878358496118065,
      "grad_norm": 0.40884220600128174,
      "learning_rate": 3.8089240277906804e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 20898816,
      "step": 12985
    },
    {
      "epoch": 0.5880622014984495,
      "grad_norm": 0.5768678188323975,
      "learning_rate": 3.808087494629535e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 20906976,
      "step": 12990
    },
    {
      "epoch": 0.5882885533850924,
      "grad_norm": 0.6175409555435181,
      "learning_rate": 3.8072507597448595e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 20914784,
      "step": 12995
    },
    {
      "epoch": 0.5885149052717354,
      "grad_norm": 0.6045204401016235,
      "learning_rate": 3.806413823265689e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 20923456,
      "step": 13000
    },
    {
      "epoch": 0.5885149052717354,
      "eval_loss": 0.154221311211586,
      "eval_runtime": 404.5658,
      "eval_samples_per_second": 97.069,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 20923456,
      "step": 13000
    },
    {
      "epoch": 0.5887412571583784,
      "grad_norm": 0.6290116310119629,
      "learning_rate": 3.805576685321089e-05,
      "loss": 0.1844,
      "num_input_tokens_seen": 20931776,
      "step": 13005
    },
    {
      "epoch": 0.5889676090450214,
      "grad_norm": 0.84920334815979,
      "learning_rate": 3.804739346040158e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 20939360,
      "step": 13010
    },
    {
      "epoch": 0.5891939609316643,
      "grad_norm": 0.44964471459388733,
      "learning_rate": 3.8039018055520234e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 20946816,
      "step": 13015
    },
    {
      "epoch": 0.5894203128183073,
      "grad_norm": 0.5036370754241943,
      "learning_rate": 3.803064063985844e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 20954560,
      "step": 13020
    },
    {
      "epoch": 0.5896466647049503,
      "grad_norm": 1.3458086252212524,
      "learning_rate": 3.802226121470811e-05,
      "loss": 0.2009,
      "num_input_tokens_seen": 20962336,
      "step": 13025
    },
    {
      "epoch": 0.5898730165915933,
      "grad_norm": 0.8629865646362305,
      "learning_rate": 3.801387978136145e-05,
      "loss": 0.1869,
      "num_input_tokens_seen": 20970080,
      "step": 13030
    },
    {
      "epoch": 0.5900993684782363,
      "grad_norm": 0.8610244989395142,
      "learning_rate": 3.800549634111099e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 20977536,
      "step": 13035
    },
    {
      "epoch": 0.5903257203648793,
      "grad_norm": 1.6062201261520386,
      "learning_rate": 3.799711089524955e-05,
      "loss": 0.2003,
      "num_input_tokens_seen": 20986208,
      "step": 13040
    },
    {
      "epoch": 0.5905520722515222,
      "grad_norm": 0.7227742671966553,
      "learning_rate": 3.7988723445070285e-05,
      "loss": 0.17,
      "num_input_tokens_seen": 20994304,
      "step": 13045
    },
    {
      "epoch": 0.5907784241381652,
      "grad_norm": 0.41410306096076965,
      "learning_rate": 3.798033399186663e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 21002944,
      "step": 13050
    },
    {
      "epoch": 0.5910047760248082,
      "grad_norm": 0.6900568008422852,
      "learning_rate": 3.797194253693237e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 21010784,
      "step": 13055
    },
    {
      "epoch": 0.5912311279114512,
      "grad_norm": 0.8037823438644409,
      "learning_rate": 3.796354908156153e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 21018784,
      "step": 13060
    },
    {
      "epoch": 0.5914574797980942,
      "grad_norm": 1.1490669250488281,
      "learning_rate": 3.795515362704853e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 21026880,
      "step": 13065
    },
    {
      "epoch": 0.591683831684737,
      "grad_norm": 1.4790438413619995,
      "learning_rate": 3.794675617468803e-05,
      "loss": 0.1928,
      "num_input_tokens_seen": 21034528,
      "step": 13070
    },
    {
      "epoch": 0.59191018357138,
      "grad_norm": 0.5237748622894287,
      "learning_rate": 3.793835672577503e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 21042400,
      "step": 13075
    },
    {
      "epoch": 0.592136535458023,
      "grad_norm": 0.6963104605674744,
      "learning_rate": 3.7929955281604826e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 21049984,
      "step": 13080
    },
    {
      "epoch": 0.592362887344666,
      "grad_norm": 0.3674826920032501,
      "learning_rate": 3.7921551843473036e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 21057792,
      "step": 13085
    },
    {
      "epoch": 0.592589239231309,
      "grad_norm": 0.39770856499671936,
      "learning_rate": 3.791314641267557e-05,
      "loss": 0.0835,
      "num_input_tokens_seen": 21065440,
      "step": 13090
    },
    {
      "epoch": 0.592815591117952,
      "grad_norm": 0.6292266845703125,
      "learning_rate": 3.790473899050864e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 21073344,
      "step": 13095
    },
    {
      "epoch": 0.5930419430045949,
      "grad_norm": 0.917560875415802,
      "learning_rate": 3.7896329578268794e-05,
      "loss": 0.1993,
      "num_input_tokens_seen": 21081280,
      "step": 13100
    },
    {
      "epoch": 0.5932682948912379,
      "grad_norm": 0.45486289262771606,
      "learning_rate": 3.7887918177252855e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 21089728,
      "step": 13105
    },
    {
      "epoch": 0.5934946467778809,
      "grad_norm": 0.8748961091041565,
      "learning_rate": 3.787950478875798e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 21098272,
      "step": 13110
    },
    {
      "epoch": 0.5937209986645239,
      "grad_norm": 0.3086812198162079,
      "learning_rate": 3.787108941408162e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 21106656,
      "step": 13115
    },
    {
      "epoch": 0.5939473505511669,
      "grad_norm": 0.5780647397041321,
      "learning_rate": 3.786267205452151e-05,
      "loss": 0.1179,
      "num_input_tokens_seen": 21114880,
      "step": 13120
    },
    {
      "epoch": 0.5941737024378099,
      "grad_norm": 0.6658218502998352,
      "learning_rate": 3.785425271137573e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 21122688,
      "step": 13125
    },
    {
      "epoch": 0.5944000543244528,
      "grad_norm": 0.9689229130744934,
      "learning_rate": 3.7845831385942655e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 21130464,
      "step": 13130
    },
    {
      "epoch": 0.5946264062110957,
      "grad_norm": 0.7219911217689514,
      "learning_rate": 3.7837408079520944e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 21139008,
      "step": 13135
    },
    {
      "epoch": 0.5948527580977387,
      "grad_norm": 1.1121355295181274,
      "learning_rate": 3.782898279340957e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 21147200,
      "step": 13140
    },
    {
      "epoch": 0.5950791099843817,
      "grad_norm": 0.31361979246139526,
      "learning_rate": 3.782055552890784e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 21155200,
      "step": 13145
    },
    {
      "epoch": 0.5953054618710247,
      "grad_norm": 0.7287588119506836,
      "learning_rate": 3.781212628731534e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 21162784,
      "step": 13150
    },
    {
      "epoch": 0.5955318137576676,
      "grad_norm": 0.7890275716781616,
      "learning_rate": 3.7803695069931946e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 21170720,
      "step": 13155
    },
    {
      "epoch": 0.5957581656443106,
      "grad_norm": 0.7870166897773743,
      "learning_rate": 3.779526187805789e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 21178368,
      "step": 13160
    },
    {
      "epoch": 0.5959845175309536,
      "grad_norm": 0.7249295115470886,
      "learning_rate": 3.778682671299364e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 21186464,
      "step": 13165
    },
    {
      "epoch": 0.5962108694175966,
      "grad_norm": 0.436540812253952,
      "learning_rate": 3.777838957604003e-05,
      "loss": 0.1214,
      "num_input_tokens_seen": 21195424,
      "step": 13170
    },
    {
      "epoch": 0.5964372213042396,
      "grad_norm": 0.5583291053771973,
      "learning_rate": 3.776995046849816e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 21203168,
      "step": 13175
    },
    {
      "epoch": 0.5966635731908826,
      "grad_norm": 0.6312472224235535,
      "learning_rate": 3.776150939166945e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 21210944,
      "step": 13180
    },
    {
      "epoch": 0.5968899250775255,
      "grad_norm": 0.9323354363441467,
      "learning_rate": 3.775306634685562e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 21218592,
      "step": 13185
    },
    {
      "epoch": 0.5971162769641685,
      "grad_norm": 0.6193430423736572,
      "learning_rate": 3.7744621335358696e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 21226304,
      "step": 13190
    },
    {
      "epoch": 0.5973426288508115,
      "grad_norm": 1.3160736560821533,
      "learning_rate": 3.7736174358481e-05,
      "loss": 0.2079,
      "num_input_tokens_seen": 21234848,
      "step": 13195
    },
    {
      "epoch": 0.5975689807374545,
      "grad_norm": 0.4853249788284302,
      "learning_rate": 3.7727725417525175e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 21242368,
      "step": 13200
    },
    {
      "epoch": 0.5975689807374545,
      "eval_loss": 0.1537107676267624,
      "eval_runtime": 404.5627,
      "eval_samples_per_second": 97.07,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 21242368,
      "step": 13200
    },
    {
      "epoch": 0.5977953326240975,
      "grad_norm": 1.0792055130004883,
      "learning_rate": 3.771927451379414e-05,
      "loss": 0.1716,
      "num_input_tokens_seen": 21250784,
      "step": 13205
    },
    {
      "epoch": 0.5980216845107404,
      "grad_norm": 1.2739521265029907,
      "learning_rate": 3.7710821648591135e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 21258752,
      "step": 13210
    },
    {
      "epoch": 0.5982480363973833,
      "grad_norm": 1.4120800495147705,
      "learning_rate": 3.7702366823219694e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 21266880,
      "step": 13215
    },
    {
      "epoch": 0.5984743882840263,
      "grad_norm": 0.860573410987854,
      "learning_rate": 3.769391003898366e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 21274944,
      "step": 13220
    },
    {
      "epoch": 0.5987007401706693,
      "grad_norm": 1.9512492418289185,
      "learning_rate": 3.768545129718718e-05,
      "loss": 0.2205,
      "num_input_tokens_seen": 21283008,
      "step": 13225
    },
    {
      "epoch": 0.5989270920573123,
      "grad_norm": 0.3549679219722748,
      "learning_rate": 3.7676990599134686e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 21290816,
      "step": 13230
    },
    {
      "epoch": 0.5991534439439553,
      "grad_norm": 1.2323596477508545,
      "learning_rate": 3.766852794613095e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 21298752,
      "step": 13235
    },
    {
      "epoch": 0.5993797958305982,
      "grad_norm": 0.5464047193527222,
      "learning_rate": 3.766006333948099e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 21306720,
      "step": 13240
    },
    {
      "epoch": 0.5996061477172412,
      "grad_norm": 0.8885177969932556,
      "learning_rate": 3.765159678049017e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 21314624,
      "step": 13245
    },
    {
      "epoch": 0.5998324996038842,
      "grad_norm": 0.5116797685623169,
      "learning_rate": 3.7643128270464134e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 21322176,
      "step": 13250
    },
    {
      "epoch": 0.6000588514905272,
      "grad_norm": 0.6889097690582275,
      "learning_rate": 3.763465781070884e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 21329888,
      "step": 13255
    },
    {
      "epoch": 0.6002852033771702,
      "grad_norm": 1.2388851642608643,
      "learning_rate": 3.762618540253052e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 21337920,
      "step": 13260
    },
    {
      "epoch": 0.6005115552638132,
      "grad_norm": 0.36218976974487305,
      "learning_rate": 3.761771104723576e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 21345536,
      "step": 13265
    },
    {
      "epoch": 0.600737907150456,
      "grad_norm": 0.36022236943244934,
      "learning_rate": 3.7609234746131386e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 21353312,
      "step": 13270
    },
    {
      "epoch": 0.600964259037099,
      "grad_norm": 0.4399649500846863,
      "learning_rate": 3.7600756500524556e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 21361120,
      "step": 13275
    },
    {
      "epoch": 0.601190610923742,
      "grad_norm": 0.5353163480758667,
      "learning_rate": 3.759227631172271e-05,
      "loss": 0.2297,
      "num_input_tokens_seen": 21369504,
      "step": 13280
    },
    {
      "epoch": 0.601416962810385,
      "grad_norm": 0.7520326972007751,
      "learning_rate": 3.758379418103363e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 21377248,
      "step": 13285
    },
    {
      "epoch": 0.601643314697028,
      "grad_norm": 1.0932166576385498,
      "learning_rate": 3.757531010976534e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 21385408,
      "step": 13290
    },
    {
      "epoch": 0.601869666583671,
      "grad_norm": 0.5924074649810791,
      "learning_rate": 3.75668240992262e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 21393568,
      "step": 13295
    },
    {
      "epoch": 0.6020960184703139,
      "grad_norm": 0.34790146350860596,
      "learning_rate": 3.7558336150724865e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 21402912,
      "step": 13300
    },
    {
      "epoch": 0.6023223703569569,
      "grad_norm": 0.4100446105003357,
      "learning_rate": 3.754984626557028e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 21411008,
      "step": 13305
    },
    {
      "epoch": 0.6025487222435999,
      "grad_norm": 1.1921164989471436,
      "learning_rate": 3.754135444507168e-05,
      "loss": 0.1622,
      "num_input_tokens_seen": 21418848,
      "step": 13310
    },
    {
      "epoch": 0.6027750741302429,
      "grad_norm": 0.8402495384216309,
      "learning_rate": 3.753286069053863e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 21426752,
      "step": 13315
    },
    {
      "epoch": 0.6030014260168859,
      "grad_norm": 0.442383736371994,
      "learning_rate": 3.7524365003280945e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 21434688,
      "step": 13320
    },
    {
      "epoch": 0.6032277779035288,
      "grad_norm": 0.8751981854438782,
      "learning_rate": 3.75158673846088e-05,
      "loss": 0.205,
      "num_input_tokens_seen": 21442208,
      "step": 13325
    },
    {
      "epoch": 0.6034541297901718,
      "grad_norm": 1.0805203914642334,
      "learning_rate": 3.750736783583262e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 21450336,
      "step": 13330
    },
    {
      "epoch": 0.6036804816768148,
      "grad_norm": 0.6093159914016724,
      "learning_rate": 3.7498866358263144e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 21458336,
      "step": 13335
    },
    {
      "epoch": 0.6039068335634578,
      "grad_norm": 0.47440141439437866,
      "learning_rate": 3.74903629532114e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 21465888,
      "step": 13340
    },
    {
      "epoch": 0.6041331854501008,
      "grad_norm": 0.5622010231018066,
      "learning_rate": 3.748185762198873e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 21473408,
      "step": 13345
    },
    {
      "epoch": 0.6043595373367437,
      "grad_norm": 1.2381713390350342,
      "learning_rate": 3.747335036590676e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 21481312,
      "step": 13350
    },
    {
      "epoch": 0.6045858892233866,
      "grad_norm": 0.8316443562507629,
      "learning_rate": 3.7464841186277405e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 21489312,
      "step": 13355
    },
    {
      "epoch": 0.6048122411100296,
      "grad_norm": 0.7132692933082581,
      "learning_rate": 3.7456330084412896e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 21497664,
      "step": 13360
    },
    {
      "epoch": 0.6050385929966726,
      "grad_norm": 1.112503170967102,
      "learning_rate": 3.744781706162576e-05,
      "loss": 0.1931,
      "num_input_tokens_seen": 21506144,
      "step": 13365
    },
    {
      "epoch": 0.6052649448833156,
      "grad_norm": 0.6381329894065857,
      "learning_rate": 3.743930211922879e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 21514112,
      "step": 13370
    },
    {
      "epoch": 0.6054912967699586,
      "grad_norm": 0.39052098989486694,
      "learning_rate": 3.743078525853513e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 21522112,
      "step": 13375
    },
    {
      "epoch": 0.6057176486566016,
      "grad_norm": 0.5370749235153198,
      "learning_rate": 3.7422266480858154e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 21529568,
      "step": 13380
    },
    {
      "epoch": 0.6059440005432445,
      "grad_norm": 0.6089954972267151,
      "learning_rate": 3.741374578751158e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 21538080,
      "step": 13385
    },
    {
      "epoch": 0.6061703524298875,
      "grad_norm": 0.6828940510749817,
      "learning_rate": 3.740522317980941e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 21545824,
      "step": 13390
    },
    {
      "epoch": 0.6063967043165305,
      "grad_norm": 0.810990571975708,
      "learning_rate": 3.739669865906593e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 21553696,
      "step": 13395
    },
    {
      "epoch": 0.6066230562031735,
      "grad_norm": 0.55585116147995,
      "learning_rate": 3.738817222659573e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 21561952,
      "step": 13400
    },
    {
      "epoch": 0.6066230562031735,
      "eval_loss": 0.15437451004981995,
      "eval_runtime": 404.2838,
      "eval_samples_per_second": 97.137,
      "eval_steps_per_second": 24.285,
      "num_input_tokens_seen": 21561952,
      "step": 13400
    },
    {
      "epoch": 0.6068494080898165,
      "grad_norm": 0.49119219183921814,
      "learning_rate": 3.73796438837137e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 21569792,
      "step": 13405
    },
    {
      "epoch": 0.6070757599764594,
      "grad_norm": 0.4077080190181732,
      "learning_rate": 3.7371113631735e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 21578240,
      "step": 13410
    },
    {
      "epoch": 0.6073021118631023,
      "grad_norm": 1.1830374002456665,
      "learning_rate": 3.736258147197512e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 21585888,
      "step": 13415
    },
    {
      "epoch": 0.6075284637497453,
      "grad_norm": 0.8159908652305603,
      "learning_rate": 3.735404740574981e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 21593600,
      "step": 13420
    },
    {
      "epoch": 0.6077548156363883,
      "grad_norm": 1.0020840167999268,
      "learning_rate": 3.7345511434375145e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 21601280,
      "step": 13425
    },
    {
      "epoch": 0.6079811675230313,
      "grad_norm": 0.8718625903129578,
      "learning_rate": 3.733697355916748e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 21609408,
      "step": 13430
    },
    {
      "epoch": 0.6082075194096743,
      "grad_norm": 0.44604653120040894,
      "learning_rate": 3.732843378144345e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 21616896,
      "step": 13435
    },
    {
      "epoch": 0.6084338712963172,
      "grad_norm": 0.9308622479438782,
      "learning_rate": 3.7319892102519995e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 21624960,
      "step": 13440
    },
    {
      "epoch": 0.6086602231829602,
      "grad_norm": 1.1131364107131958,
      "learning_rate": 3.731134852371436e-05,
      "loss": 0.1973,
      "num_input_tokens_seen": 21633312,
      "step": 13445
    },
    {
      "epoch": 0.6088865750696032,
      "grad_norm": 1.087507963180542,
      "learning_rate": 3.730280304634408e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 21640608,
      "step": 13450
    },
    {
      "epoch": 0.6091129269562462,
      "grad_norm": 0.3761869966983795,
      "learning_rate": 3.729425567172696e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 21648928,
      "step": 13455
    },
    {
      "epoch": 0.6093392788428892,
      "grad_norm": 0.43153175711631775,
      "learning_rate": 3.728570640118111e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 21658144,
      "step": 13460
    },
    {
      "epoch": 0.6095656307295322,
      "grad_norm": 0.32734912633895874,
      "learning_rate": 3.727715523602494e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 21666048,
      "step": 13465
    },
    {
      "epoch": 0.6097919826161751,
      "grad_norm": 1.4712406396865845,
      "learning_rate": 3.726860217757715e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 21673536,
      "step": 13470
    },
    {
      "epoch": 0.6100183345028181,
      "grad_norm": 0.8679699897766113,
      "learning_rate": 3.726004722715673e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 21681344,
      "step": 13475
    },
    {
      "epoch": 0.610244686389461,
      "grad_norm": 0.7336258292198181,
      "learning_rate": 3.725149038608296e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 21689440,
      "step": 13480
    },
    {
      "epoch": 0.610471038276104,
      "grad_norm": 0.6356282234191895,
      "learning_rate": 3.7242931655675404e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 21697120,
      "step": 13485
    },
    {
      "epoch": 0.610697390162747,
      "grad_norm": 0.8108717799186707,
      "learning_rate": 3.7234371037253937e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 21704992,
      "step": 13490
    },
    {
      "epoch": 0.6109237420493899,
      "grad_norm": 0.6737693548202515,
      "learning_rate": 3.7225808532138705e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 21713632,
      "step": 13495
    },
    {
      "epoch": 0.6111500939360329,
      "grad_norm": 0.4840019643306732,
      "learning_rate": 3.721724414165016e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 21721056,
      "step": 13500
    },
    {
      "epoch": 0.6113764458226759,
      "grad_norm": 0.5399805903434753,
      "learning_rate": 3.720867786710904e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 21729088,
      "step": 13505
    },
    {
      "epoch": 0.6116027977093189,
      "grad_norm": 0.7658386826515198,
      "learning_rate": 3.7200109709836366e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 21737344,
      "step": 13510
    },
    {
      "epoch": 0.6118291495959619,
      "grad_norm": 1.353534460067749,
      "learning_rate": 3.7191539671153465e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 21745312,
      "step": 13515
    },
    {
      "epoch": 0.6120555014826049,
      "grad_norm": 0.47510823607444763,
      "learning_rate": 3.718296775238193e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 21753696,
      "step": 13520
    },
    {
      "epoch": 0.6122818533692478,
      "grad_norm": 0.745053768157959,
      "learning_rate": 3.7174393954843675e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 21761312,
      "step": 13525
    },
    {
      "epoch": 0.6125082052558908,
      "grad_norm": 0.5950844883918762,
      "learning_rate": 3.716581827986087e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 21769504,
      "step": 13530
    },
    {
      "epoch": 0.6127345571425338,
      "grad_norm": 0.6599022150039673,
      "learning_rate": 3.7157240728756004e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 21777280,
      "step": 13535
    },
    {
      "epoch": 0.6129609090291768,
      "grad_norm": 1.0608956813812256,
      "learning_rate": 3.714866130285184e-05,
      "loss": 0.1864,
      "num_input_tokens_seen": 21785728,
      "step": 13540
    },
    {
      "epoch": 0.6131872609158198,
      "grad_norm": 0.8215600848197937,
      "learning_rate": 3.714008000347143e-05,
      "loss": 0.125,
      "num_input_tokens_seen": 21793824,
      "step": 13545
    },
    {
      "epoch": 0.6134136128024628,
      "grad_norm": 0.3852002024650574,
      "learning_rate": 3.7131496831938126e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 21801600,
      "step": 13550
    },
    {
      "epoch": 0.6136399646891056,
      "grad_norm": 0.7851988077163696,
      "learning_rate": 3.7122911789575565e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 21809600,
      "step": 13555
    },
    {
      "epoch": 0.6138663165757486,
      "grad_norm": 0.833954930305481,
      "learning_rate": 3.711432487770765e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 21817760,
      "step": 13560
    },
    {
      "epoch": 0.6140926684623916,
      "grad_norm": 0.6210547089576721,
      "learning_rate": 3.710573609765861e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 21825632,
      "step": 13565
    },
    {
      "epoch": 0.6143190203490346,
      "grad_norm": 0.7946431040763855,
      "learning_rate": 3.709714545075292e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 21833504,
      "step": 13570
    },
    {
      "epoch": 0.6145453722356776,
      "grad_norm": 0.33486348390579224,
      "learning_rate": 3.708855293831538e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 21841984,
      "step": 13575
    },
    {
      "epoch": 0.6147717241223205,
      "grad_norm": 0.874879002571106,
      "learning_rate": 3.707995856167107e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 21849952,
      "step": 13580
    },
    {
      "epoch": 0.6149980760089635,
      "grad_norm": 1.8883715867996216,
      "learning_rate": 3.707136232214534e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 21857920,
      "step": 13585
    },
    {
      "epoch": 0.6152244278956065,
      "grad_norm": 0.864851713180542,
      "learning_rate": 3.7062764221063844e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 21866240,
      "step": 13590
    },
    {
      "epoch": 0.6154507797822495,
      "grad_norm": 0.5786483883857727,
      "learning_rate": 3.705416425975252e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 21873856,
      "step": 13595
    },
    {
      "epoch": 0.6156771316688925,
      "grad_norm": 0.6729029417037964,
      "learning_rate": 3.704556243953758e-05,
      "loss": 0.1715,
      "num_input_tokens_seen": 21881856,
      "step": 13600
    },
    {
      "epoch": 0.6156771316688925,
      "eval_loss": 0.15275973081588745,
      "eval_runtime": 403.8687,
      "eval_samples_per_second": 97.237,
      "eval_steps_per_second": 24.31,
      "num_input_tokens_seen": 21881856,
      "step": 13600
    },
    {
      "epoch": 0.6159034835555355,
      "grad_norm": 0.8959065079689026,
      "learning_rate": 3.7036958761745535e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 21890400,
      "step": 13605
    },
    {
      "epoch": 0.6161298354421784,
      "grad_norm": 0.6924241185188293,
      "learning_rate": 3.702835322770318e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 21898144,
      "step": 13610
    },
    {
      "epoch": 0.6163561873288214,
      "grad_norm": 0.5575647354125977,
      "learning_rate": 3.701974583873761e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 21906688,
      "step": 13615
    },
    {
      "epoch": 0.6165825392154644,
      "grad_norm": 0.5683234333992004,
      "learning_rate": 3.701113659617618e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 21914560,
      "step": 13620
    },
    {
      "epoch": 0.6168088911021073,
      "grad_norm": 0.5781339406967163,
      "learning_rate": 3.7002525501346535e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 21922368,
      "step": 13625
    },
    {
      "epoch": 0.6170352429887503,
      "grad_norm": 0.5243660807609558,
      "learning_rate": 3.699391255557664e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 21930976,
      "step": 13630
    },
    {
      "epoch": 0.6172615948753933,
      "grad_norm": 0.7609244585037231,
      "learning_rate": 3.69852977601947e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 21939488,
      "step": 13635
    },
    {
      "epoch": 0.6174879467620362,
      "grad_norm": 0.919364869594574,
      "learning_rate": 3.697668111652922e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 21947072,
      "step": 13640
    },
    {
      "epoch": 0.6177142986486792,
      "grad_norm": 0.6567410230636597,
      "learning_rate": 3.6968062625909005e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 21955168,
      "step": 13645
    },
    {
      "epoch": 0.6179406505353222,
      "grad_norm": 1.627451777458191,
      "learning_rate": 3.6959442289663135e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 21963872,
      "step": 13650
    },
    {
      "epoch": 0.6181670024219652,
      "grad_norm": 0.6699845790863037,
      "learning_rate": 3.695082010912098e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 21971264,
      "step": 13655
    },
    {
      "epoch": 0.6183933543086082,
      "grad_norm": 1.0919864177703857,
      "learning_rate": 3.694219608561217e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 21979616,
      "step": 13660
    },
    {
      "epoch": 0.6186197061952511,
      "grad_norm": 0.4012785851955414,
      "learning_rate": 3.693357022046665e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 21987904,
      "step": 13665
    },
    {
      "epoch": 0.6188460580818941,
      "grad_norm": 0.9468543529510498,
      "learning_rate": 3.6924942515014644e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 21995968,
      "step": 13670
    },
    {
      "epoch": 0.6190724099685371,
      "grad_norm": 0.5611370205879211,
      "learning_rate": 3.691631297058664e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 22004800,
      "step": 13675
    },
    {
      "epoch": 0.6192987618551801,
      "grad_norm": 0.9715312719345093,
      "learning_rate": 3.6907681588513424e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 22012320,
      "step": 13680
    },
    {
      "epoch": 0.6195251137418231,
      "grad_norm": 0.8706628084182739,
      "learning_rate": 3.689904837012606e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 22020416,
      "step": 13685
    },
    {
      "epoch": 0.6197514656284661,
      "grad_norm": 0.4105319380760193,
      "learning_rate": 3.689041331675591e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 22028320,
      "step": 13690
    },
    {
      "epoch": 0.6199778175151089,
      "grad_norm": 0.5681552886962891,
      "learning_rate": 3.688177642973461e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 22037152,
      "step": 13695
    },
    {
      "epoch": 0.6202041694017519,
      "grad_norm": 0.5706701874732971,
      "learning_rate": 3.687313771039406e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 22045568,
      "step": 13700
    },
    {
      "epoch": 0.6204305212883949,
      "grad_norm": 1.4123106002807617,
      "learning_rate": 3.686449716006647e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 22053632,
      "step": 13705
    },
    {
      "epoch": 0.6206568731750379,
      "grad_norm": 0.4789370894432068,
      "learning_rate": 3.685585478008432e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 22062400,
      "step": 13710
    },
    {
      "epoch": 0.6208832250616809,
      "grad_norm": 1.143530011177063,
      "learning_rate": 3.6847210571780364e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 22069984,
      "step": 13715
    },
    {
      "epoch": 0.6211095769483239,
      "grad_norm": 0.30519899725914,
      "learning_rate": 3.683856453648767e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 22077952,
      "step": 13720
    },
    {
      "epoch": 0.6213359288349668,
      "grad_norm": 1.1253563165664673,
      "learning_rate": 3.682991667553954e-05,
      "loss": 0.1743,
      "num_input_tokens_seen": 22086368,
      "step": 13725
    },
    {
      "epoch": 0.6215622807216098,
      "grad_norm": 0.30595317482948303,
      "learning_rate": 3.6821266990269606e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 22094688,
      "step": 13730
    },
    {
      "epoch": 0.6217886326082528,
      "grad_norm": 0.8070888519287109,
      "learning_rate": 3.681261548201174e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 22102912,
      "step": 13735
    },
    {
      "epoch": 0.6220149844948958,
      "grad_norm": 0.7984886765480042,
      "learning_rate": 3.6803962152100125e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 22110784,
      "step": 13740
    },
    {
      "epoch": 0.6222413363815388,
      "grad_norm": 1.1423999071121216,
      "learning_rate": 3.67953070018692e-05,
      "loss": 0.1555,
      "num_input_tokens_seen": 22118688,
      "step": 13745
    },
    {
      "epoch": 0.6224676882681817,
      "grad_norm": 0.577754557132721,
      "learning_rate": 3.678665003265371e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 22126688,
      "step": 13750
    },
    {
      "epoch": 0.6226940401548247,
      "grad_norm": 0.7949138283729553,
      "learning_rate": 3.677799124578867e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 22134368,
      "step": 13755
    },
    {
      "epoch": 0.6229203920414677,
      "grad_norm": 0.7274842262268066,
      "learning_rate": 3.676933064260937e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 22142752,
      "step": 13760
    },
    {
      "epoch": 0.6231467439281106,
      "grad_norm": 0.6362159848213196,
      "learning_rate": 3.6760668224451365e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 22150752,
      "step": 13765
    },
    {
      "epoch": 0.6233730958147536,
      "grad_norm": 0.4775896370410919,
      "learning_rate": 3.675200399265054e-05,
      "loss": 0.1693,
      "num_input_tokens_seen": 22159392,
      "step": 13770
    },
    {
      "epoch": 0.6235994477013966,
      "grad_norm": 0.9066118597984314,
      "learning_rate": 3.6743337948543014e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 22168032,
      "step": 13775
    },
    {
      "epoch": 0.6238257995880395,
      "grad_norm": 0.7185283899307251,
      "learning_rate": 3.6734670093465204e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 22176352,
      "step": 13780
    },
    {
      "epoch": 0.6240521514746825,
      "grad_norm": 0.3169030249118805,
      "learning_rate": 3.672600042875379e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 22184288,
      "step": 13785
    },
    {
      "epoch": 0.6242785033613255,
      "grad_norm": 0.526171088218689,
      "learning_rate": 3.671732895574575e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 22193632,
      "step": 13790
    },
    {
      "epoch": 0.6245048552479685,
      "grad_norm": 0.6397806406021118,
      "learning_rate": 3.670865567577834e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 22202528,
      "step": 13795
    },
    {
      "epoch": 0.6247312071346115,
      "grad_norm": 0.5662399530410767,
      "learning_rate": 3.669998059018909e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 22210080,
      "step": 13800
    },
    {
      "epoch": 0.6247312071346115,
      "eval_loss": 0.15467651188373566,
      "eval_runtime": 404.9114,
      "eval_samples_per_second": 96.987,
      "eval_steps_per_second": 24.247,
      "num_input_tokens_seen": 22210080,
      "step": 13800
    },
    {
      "epoch": 0.6249575590212545,
      "grad_norm": 0.4167740046977997,
      "learning_rate": 3.6691303700315796e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 22218048,
      "step": 13805
    },
    {
      "epoch": 0.6251839109078974,
      "grad_norm": 0.6071748733520508,
      "learning_rate": 3.668262500749655e-05,
      "loss": 0.202,
      "num_input_tokens_seen": 22225856,
      "step": 13810
    },
    {
      "epoch": 0.6254102627945404,
      "grad_norm": 0.2729750871658325,
      "learning_rate": 3.667394451306971e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 22234048,
      "step": 13815
    },
    {
      "epoch": 0.6256366146811834,
      "grad_norm": 0.558067798614502,
      "learning_rate": 3.666526221837393e-05,
      "loss": 0.1912,
      "num_input_tokens_seen": 22241408,
      "step": 13820
    },
    {
      "epoch": 0.6258629665678264,
      "grad_norm": 0.5470388531684875,
      "learning_rate": 3.665657812474812e-05,
      "loss": 0.2187,
      "num_input_tokens_seen": 22249376,
      "step": 13825
    },
    {
      "epoch": 0.6260893184544694,
      "grad_norm": 0.3378750681877136,
      "learning_rate": 3.664789223353147e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 22257600,
      "step": 13830
    },
    {
      "epoch": 0.6263156703411122,
      "grad_norm": 0.9989092946052551,
      "learning_rate": 3.663920454606347e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 22265248,
      "step": 13835
    },
    {
      "epoch": 0.6265420222277552,
      "grad_norm": 1.082216739654541,
      "learning_rate": 3.6630515063683856e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 22273440,
      "step": 13840
    },
    {
      "epoch": 0.6267683741143982,
      "grad_norm": 0.9324672818183899,
      "learning_rate": 3.662182378773267e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 22281312,
      "step": 13845
    },
    {
      "epoch": 0.6269947260010412,
      "grad_norm": 0.4536956548690796,
      "learning_rate": 3.66131307195502e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 22288992,
      "step": 13850
    },
    {
      "epoch": 0.6272210778876842,
      "grad_norm": 0.35846030712127686,
      "learning_rate": 3.6604435860477034e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 22296320,
      "step": 13855
    },
    {
      "epoch": 0.6274474297743272,
      "grad_norm": 0.5330339670181274,
      "learning_rate": 3.6595739211854025e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 22304512,
      "step": 13860
    },
    {
      "epoch": 0.6276737816609701,
      "grad_norm": 0.8817346096038818,
      "learning_rate": 3.658704077502231e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 22313536,
      "step": 13865
    },
    {
      "epoch": 0.6279001335476131,
      "grad_norm": 0.45734068751335144,
      "learning_rate": 3.65783405513233e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 22321792,
      "step": 13870
    },
    {
      "epoch": 0.6281264854342561,
      "grad_norm": 1.216583490371704,
      "learning_rate": 3.656963854209867e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 22330048,
      "step": 13875
    },
    {
      "epoch": 0.6283528373208991,
      "grad_norm": 0.6865096092224121,
      "learning_rate": 3.656093474869038e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 22337312,
      "step": 13880
    },
    {
      "epoch": 0.6285791892075421,
      "grad_norm": 0.35343727469444275,
      "learning_rate": 3.655222917244068e-05,
      "loss": 0.1119,
      "num_input_tokens_seen": 22345312,
      "step": 13885
    },
    {
      "epoch": 0.6288055410941851,
      "grad_norm": 0.9225030541419983,
      "learning_rate": 3.6543521814692054e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 22353376,
      "step": 13890
    },
    {
      "epoch": 0.629031892980828,
      "grad_norm": 0.5701672434806824,
      "learning_rate": 3.653481267678731e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 22361632,
      "step": 13895
    },
    {
      "epoch": 0.629258244867471,
      "grad_norm": 0.4191988706588745,
      "learning_rate": 3.652610176006949e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 22369728,
      "step": 13900
    },
    {
      "epoch": 0.629484596754114,
      "grad_norm": 1.1976091861724854,
      "learning_rate": 3.6517389065881925e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 22377632,
      "step": 13905
    },
    {
      "epoch": 0.6297109486407569,
      "grad_norm": 0.8459890484809875,
      "learning_rate": 3.650867459556824e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 22385184,
      "step": 13910
    },
    {
      "epoch": 0.6299373005273999,
      "grad_norm": 0.5734026432037354,
      "learning_rate": 3.64999583504723e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 22394336,
      "step": 13915
    },
    {
      "epoch": 0.6301636524140428,
      "grad_norm": 0.9774382710456848,
      "learning_rate": 3.649124033193827e-05,
      "loss": 0.1904,
      "num_input_tokens_seen": 22401952,
      "step": 13920
    },
    {
      "epoch": 0.6303900043006858,
      "grad_norm": 0.439929336309433,
      "learning_rate": 3.648252054131057e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 22410176,
      "step": 13925
    },
    {
      "epoch": 0.6306163561873288,
      "grad_norm": 0.4602413773536682,
      "learning_rate": 3.647379897993391e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 22418304,
      "step": 13930
    },
    {
      "epoch": 0.6308427080739718,
      "grad_norm": 0.966065526008606,
      "learning_rate": 3.646507564915325e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 22425920,
      "step": 13935
    },
    {
      "epoch": 0.6310690599606148,
      "grad_norm": 0.8316280245780945,
      "learning_rate": 3.645635055031385e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 22433824,
      "step": 13940
    },
    {
      "epoch": 0.6312954118472578,
      "grad_norm": 1.1011534929275513,
      "learning_rate": 3.6447623684761224e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 22442112,
      "step": 13945
    },
    {
      "epoch": 0.6315217637339007,
      "grad_norm": 0.3196682929992676,
      "learning_rate": 3.643889505384117e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 22450304,
      "step": 13950
    },
    {
      "epoch": 0.6317481156205437,
      "grad_norm": 1.072729468345642,
      "learning_rate": 3.6430164658899744e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 22457792,
      "step": 13955
    },
    {
      "epoch": 0.6319744675071867,
      "grad_norm": 0.4928816556930542,
      "learning_rate": 3.642143250128329e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 22465472,
      "step": 13960
    },
    {
      "epoch": 0.6322008193938297,
      "grad_norm": 0.7033950686454773,
      "learning_rate": 3.641269858233841e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 22473408,
      "step": 13965
    },
    {
      "epoch": 0.6324271712804727,
      "grad_norm": 0.6926734447479248,
      "learning_rate": 3.640396290341199e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 22481632,
      "step": 13970
    },
    {
      "epoch": 0.6326535231671157,
      "grad_norm": 0.5072685480117798,
      "learning_rate": 3.639522546585118e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 22489696,
      "step": 13975
    },
    {
      "epoch": 0.6328798750537585,
      "grad_norm": 0.5816423296928406,
      "learning_rate": 3.6386486271003404e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 22497248,
      "step": 13980
    },
    {
      "epoch": 0.6331062269404015,
      "grad_norm": 0.6631471514701843,
      "learning_rate": 3.6377745320216346e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 22505312,
      "step": 13985
    },
    {
      "epoch": 0.6333325788270445,
      "grad_norm": 0.48956677317619324,
      "learning_rate": 3.636900261483798e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 22513888,
      "step": 13990
    },
    {
      "epoch": 0.6335589307136875,
      "grad_norm": 1.2596800327301025,
      "learning_rate": 3.636025815621654e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 22521600,
      "step": 13995
    },
    {
      "epoch": 0.6337852826003305,
      "grad_norm": 0.5249874591827393,
      "learning_rate": 3.635151194570054e-05,
      "loss": 0.1782,
      "num_input_tokens_seen": 22529216,
      "step": 14000
    },
    {
      "epoch": 0.6337852826003305,
      "eval_loss": 0.15264566242694855,
      "eval_runtime": 404.9749,
      "eval_samples_per_second": 96.971,
      "eval_steps_per_second": 24.243,
      "num_input_tokens_seen": 22529216,
      "step": 14000
    },
    {
      "epoch": 0.6340116344869734,
      "grad_norm": 0.9147670269012451,
      "learning_rate": 3.634276398463873e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 22537184,
      "step": 14005
    },
    {
      "epoch": 0.6342379863736164,
      "grad_norm": 0.518096923828125,
      "learning_rate": 3.633401427438018e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 22544768,
      "step": 14010
    },
    {
      "epoch": 0.6344643382602594,
      "grad_norm": 0.7700987458229065,
      "learning_rate": 3.63252628162742e-05,
      "loss": 0.1777,
      "num_input_tokens_seen": 22552160,
      "step": 14015
    },
    {
      "epoch": 0.6346906901469024,
      "grad_norm": 0.7405732870101929,
      "learning_rate": 3.6316509611670364e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 22560832,
      "step": 14020
    },
    {
      "epoch": 0.6349170420335454,
      "grad_norm": 0.43886953592300415,
      "learning_rate": 3.630775466191854e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 22568576,
      "step": 14025
    },
    {
      "epoch": 0.6351433939201884,
      "grad_norm": 0.5210665464401245,
      "learning_rate": 3.629899796836884e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 22576736,
      "step": 14030
    },
    {
      "epoch": 0.6353697458068313,
      "grad_norm": 0.7973574995994568,
      "learning_rate": 3.6290239532371666e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 22584736,
      "step": 14035
    },
    {
      "epoch": 0.6355960976934742,
      "grad_norm": 0.593773365020752,
      "learning_rate": 3.628147935527767e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 22592320,
      "step": 14040
    },
    {
      "epoch": 0.6358224495801172,
      "grad_norm": 0.47017961740493774,
      "learning_rate": 3.627271743843779e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 22600640,
      "step": 14045
    },
    {
      "epoch": 0.6360488014667602,
      "grad_norm": 0.9404605031013489,
      "learning_rate": 3.626395378320321e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 22608864,
      "step": 14050
    },
    {
      "epoch": 0.6362751533534032,
      "grad_norm": 0.35374715924263,
      "learning_rate": 3.625518839092541e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 22616384,
      "step": 14055
    },
    {
      "epoch": 0.6365015052400462,
      "grad_norm": 0.5399088263511658,
      "learning_rate": 3.624642126295612e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 22624544,
      "step": 14060
    },
    {
      "epoch": 0.6367278571266891,
      "grad_norm": 0.3934209942817688,
      "learning_rate": 3.6237652400647345e-05,
      "loss": 0.1126,
      "num_input_tokens_seen": 22633152,
      "step": 14065
    },
    {
      "epoch": 0.6369542090133321,
      "grad_norm": 0.7498579025268555,
      "learning_rate": 3.622888180535134e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 22641248,
      "step": 14070
    },
    {
      "epoch": 0.6371805608999751,
      "grad_norm": 0.3609817624092102,
      "learning_rate": 3.6220109478420655e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 22649152,
      "step": 14075
    },
    {
      "epoch": 0.6374069127866181,
      "grad_norm": 0.5616255402565002,
      "learning_rate": 3.6211335421208084e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 22657248,
      "step": 14080
    },
    {
      "epoch": 0.6376332646732611,
      "grad_norm": 0.5517336130142212,
      "learning_rate": 3.62025596350667e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 22664672,
      "step": 14085
    },
    {
      "epoch": 0.637859616559904,
      "grad_norm": 1.4696515798568726,
      "learning_rate": 3.619378212134984e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 22672512,
      "step": 14090
    },
    {
      "epoch": 0.638085968446547,
      "grad_norm": 0.7564826607704163,
      "learning_rate": 3.618500288141111e-05,
      "loss": 0.1481,
      "num_input_tokens_seen": 22680608,
      "step": 14095
    },
    {
      "epoch": 0.63831232033319,
      "grad_norm": 0.44993552565574646,
      "learning_rate": 3.617622191660438e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 22688640,
      "step": 14100
    },
    {
      "epoch": 0.638538672219833,
      "grad_norm": 0.3449820280075073,
      "learning_rate": 3.616743922828377e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 22696224,
      "step": 14105
    },
    {
      "epoch": 0.638765024106476,
      "grad_norm": 0.7739779353141785,
      "learning_rate": 3.615865481780371e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 22704288,
      "step": 14110
    },
    {
      "epoch": 0.638991375993119,
      "grad_norm": 0.6234502196311951,
      "learning_rate": 3.614986868651883e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 22711840,
      "step": 14115
    },
    {
      "epoch": 0.6392177278797618,
      "grad_norm": 1.2313330173492432,
      "learning_rate": 3.614108083578409e-05,
      "loss": 0.1796,
      "num_input_tokens_seen": 22719904,
      "step": 14120
    },
    {
      "epoch": 0.6394440797664048,
      "grad_norm": 0.6425874829292297,
      "learning_rate": 3.613229126695467e-05,
      "loss": 0.1263,
      "num_input_tokens_seen": 22727712,
      "step": 14125
    },
    {
      "epoch": 0.6396704316530478,
      "grad_norm": 0.41348496079444885,
      "learning_rate": 3.612349998138605e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 22736000,
      "step": 14130
    },
    {
      "epoch": 0.6398967835396908,
      "grad_norm": 0.896803617477417,
      "learning_rate": 3.6114706980433946e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 22743840,
      "step": 14135
    },
    {
      "epoch": 0.6401231354263338,
      "grad_norm": 0.36663123965263367,
      "learning_rate": 3.610591226545435e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 22751840,
      "step": 14140
    },
    {
      "epoch": 0.6403494873129768,
      "grad_norm": 0.770751953125,
      "learning_rate": 3.6097115837803505e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 22759392,
      "step": 14145
    },
    {
      "epoch": 0.6405758391996197,
      "grad_norm": 0.3194563388824463,
      "learning_rate": 3.608831769883795e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 22768000,
      "step": 14150
    },
    {
      "epoch": 0.6408021910862627,
      "grad_norm": 0.5102068781852722,
      "learning_rate": 3.607951784991446e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 22776160,
      "step": 14155
    },
    {
      "epoch": 0.6410285429729057,
      "grad_norm": 0.7700333595275879,
      "learning_rate": 3.6070716292390085e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 22784736,
      "step": 14160
    },
    {
      "epoch": 0.6412548948595487,
      "grad_norm": 0.45870059728622437,
      "learning_rate": 3.606191302762213e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 22793376,
      "step": 14165
    },
    {
      "epoch": 0.6414812467461917,
      "grad_norm": 0.6135222911834717,
      "learning_rate": 3.605310805696818e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 22802112,
      "step": 14170
    },
    {
      "epoch": 0.6417075986328346,
      "grad_norm": 0.44734814763069153,
      "learning_rate": 3.6044301381786067e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 22810176,
      "step": 14175
    },
    {
      "epoch": 0.6419339505194775,
      "grad_norm": 0.5630967020988464,
      "learning_rate": 3.6035493003433883e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 22817952,
      "step": 14180
    },
    {
      "epoch": 0.6421603024061205,
      "grad_norm": 0.5667471289634705,
      "learning_rate": 3.6026682923269994e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 22825792,
      "step": 14185
    },
    {
      "epoch": 0.6423866542927635,
      "grad_norm": 0.6844203472137451,
      "learning_rate": 3.6017871142653034e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 22833568,
      "step": 14190
    },
    {
      "epoch": 0.6426130061794065,
      "grad_norm": 1.5154169797897339,
      "learning_rate": 3.600905766294189e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 22841408,
      "step": 14195
    },
    {
      "epoch": 0.6428393580660495,
      "grad_norm": 0.38997676968574524,
      "learning_rate": 3.60002424854957e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 22849376,
      "step": 14200
    },
    {
      "epoch": 0.6428393580660495,
      "eval_loss": 0.15227052569389343,
      "eval_runtime": 404.5277,
      "eval_samples_per_second": 97.079,
      "eval_steps_per_second": 24.27,
      "num_input_tokens_seen": 22849376,
      "step": 14200
    },
    {
      "epoch": 0.6430657099526924,
      "grad_norm": 1.1904555559158325,
      "learning_rate": 3.5991425611673876e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 22857632,
      "step": 14205
    },
    {
      "epoch": 0.6432920618393354,
      "grad_norm": 1.1508179903030396,
      "learning_rate": 3.5982607042836105e-05,
      "loss": 0.1874,
      "num_input_tokens_seen": 22865824,
      "step": 14210
    },
    {
      "epoch": 0.6435184137259784,
      "grad_norm": 0.9820528626441956,
      "learning_rate": 3.597378678034231e-05,
      "loss": 0.1766,
      "num_input_tokens_seen": 22873760,
      "step": 14215
    },
    {
      "epoch": 0.6437447656126214,
      "grad_norm": 0.8809438347816467,
      "learning_rate": 3.596496482555269e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 22882528,
      "step": 14220
    },
    {
      "epoch": 0.6439711174992644,
      "grad_norm": 0.420665979385376,
      "learning_rate": 3.595614117982769e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 22890912,
      "step": 14225
    },
    {
      "epoch": 0.6441974693859074,
      "grad_norm": 0.6214843392372131,
      "learning_rate": 3.594731584452805e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 22899232,
      "step": 14230
    },
    {
      "epoch": 0.6444238212725503,
      "grad_norm": 0.7468595504760742,
      "learning_rate": 3.593848882101472e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 22907552,
      "step": 14235
    },
    {
      "epoch": 0.6446501731591933,
      "grad_norm": 0.7967620491981506,
      "learning_rate": 3.592966011064896e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 22915008,
      "step": 14240
    },
    {
      "epoch": 0.6448765250458363,
      "grad_norm": 1.4167249202728271,
      "learning_rate": 3.592082971479226e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 22924224,
      "step": 14245
    },
    {
      "epoch": 0.6451028769324793,
      "grad_norm": 0.503456711769104,
      "learning_rate": 3.5911997634806385e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 22932672,
      "step": 14250
    },
    {
      "epoch": 0.6453292288191222,
      "grad_norm": 0.5407741069793701,
      "learning_rate": 3.5903163872053336e-05,
      "loss": 0.1236,
      "num_input_tokens_seen": 22942048,
      "step": 14255
    },
    {
      "epoch": 0.6455555807057651,
      "grad_norm": 0.5904066562652588,
      "learning_rate": 3.58943284278954e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 22949696,
      "step": 14260
    },
    {
      "epoch": 0.6457819325924081,
      "grad_norm": 1.2136787176132202,
      "learning_rate": 3.588549130369512e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 22957664,
      "step": 14265
    },
    {
      "epoch": 0.6460082844790511,
      "grad_norm": 0.27874651551246643,
      "learning_rate": 3.5876652500815274e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 22965472,
      "step": 14270
    },
    {
      "epoch": 0.6462346363656941,
      "grad_norm": 1.0066044330596924,
      "learning_rate": 3.586781202061894e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 22973152,
      "step": 14275
    },
    {
      "epoch": 0.6464609882523371,
      "grad_norm": 0.5568544864654541,
      "learning_rate": 3.585896986446942e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 22980544,
      "step": 14280
    },
    {
      "epoch": 0.6466873401389801,
      "grad_norm": 0.41454821825027466,
      "learning_rate": 3.585012603373028e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 22988416,
      "step": 14285
    },
    {
      "epoch": 0.646913692025623,
      "grad_norm": 0.3246223032474518,
      "learning_rate": 3.584128052976535e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 22996128,
      "step": 14290
    },
    {
      "epoch": 0.647140043912266,
      "grad_norm": 1.012198805809021,
      "learning_rate": 3.5832433353938724e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 23003872,
      "step": 14295
    },
    {
      "epoch": 0.647366395798909,
      "grad_norm": 0.8480693101882935,
      "learning_rate": 3.5823584507614746e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 23011904,
      "step": 14300
    },
    {
      "epoch": 0.647592747685552,
      "grad_norm": 0.4620458483695984,
      "learning_rate": 3.581473399215802e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 23019680,
      "step": 14305
    },
    {
      "epoch": 0.647819099572195,
      "grad_norm": 0.5871486663818359,
      "learning_rate": 3.580588180893341e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 23028064,
      "step": 14310
    },
    {
      "epoch": 0.648045451458838,
      "grad_norm": 0.9981752634048462,
      "learning_rate": 3.579702795930602e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 23036000,
      "step": 14315
    },
    {
      "epoch": 0.6482718033454808,
      "grad_norm": 0.6470571160316467,
      "learning_rate": 3.578817244464125e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 23043520,
      "step": 14320
    },
    {
      "epoch": 0.6484981552321238,
      "grad_norm": 0.4979564845561981,
      "learning_rate": 3.577931526630471e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 23051488,
      "step": 14325
    },
    {
      "epoch": 0.6487245071187668,
      "grad_norm": 0.6939167976379395,
      "learning_rate": 3.577045642566229e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 23059040,
      "step": 14330
    },
    {
      "epoch": 0.6489508590054098,
      "grad_norm": 0.39030352234840393,
      "learning_rate": 3.576159592408014e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 23067008,
      "step": 14335
    },
    {
      "epoch": 0.6491772108920528,
      "grad_norm": 0.5095974802970886,
      "learning_rate": 3.575273376292466e-05,
      "loss": 0.1862,
      "num_input_tokens_seen": 23075552,
      "step": 14340
    },
    {
      "epoch": 0.6494035627786957,
      "grad_norm": 1.172201156616211,
      "learning_rate": 3.574386994356251e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 23083392,
      "step": 14345
    },
    {
      "epoch": 0.6496299146653387,
      "grad_norm": 0.4511790871620178,
      "learning_rate": 3.573500446736059e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 23091456,
      "step": 14350
    },
    {
      "epoch": 0.6498562665519817,
      "grad_norm": 0.6239246129989624,
      "learning_rate": 3.5726137335686094e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 23099168,
      "step": 14355
    },
    {
      "epoch": 0.6500826184386247,
      "grad_norm": 0.6703404784202576,
      "learning_rate": 3.571726854990642e-05,
      "loss": 0.1716,
      "num_input_tokens_seen": 23106912,
      "step": 14360
    },
    {
      "epoch": 0.6503089703252677,
      "grad_norm": 0.5550479888916016,
      "learning_rate": 3.570839811138925e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 23114400,
      "step": 14365
    },
    {
      "epoch": 0.6505353222119107,
      "grad_norm": 0.45899438858032227,
      "learning_rate": 3.569952602150252e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 23122848,
      "step": 14370
    },
    {
      "epoch": 0.6507616740985536,
      "grad_norm": 0.5725798010826111,
      "learning_rate": 3.569065228161442e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 23131072,
      "step": 14375
    },
    {
      "epoch": 0.6509880259851966,
      "grad_norm": 0.7135940790176392,
      "learning_rate": 3.5681776893093395e-05,
      "loss": 0.1484,
      "num_input_tokens_seen": 23139264,
      "step": 14380
    },
    {
      "epoch": 0.6512143778718396,
      "grad_norm": 0.7257099747657776,
      "learning_rate": 3.5672899857308134e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 23147200,
      "step": 14385
    },
    {
      "epoch": 0.6514407297584826,
      "grad_norm": 0.6746643781661987,
      "learning_rate": 3.566402117562759e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 23154912,
      "step": 14390
    },
    {
      "epoch": 0.6516670816451255,
      "grad_norm": 0.6549513339996338,
      "learning_rate": 3.565514084942097e-05,
      "loss": 0.212,
      "num_input_tokens_seen": 23163136,
      "step": 14395
    },
    {
      "epoch": 0.6518934335317685,
      "grad_norm": 0.7932209372520447,
      "learning_rate": 3.564625888005773e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 23171168,
      "step": 14400
    },
    {
      "epoch": 0.6518934335317685,
      "eval_loss": 0.15249404311180115,
      "eval_runtime": 404.5602,
      "eval_samples_per_second": 97.071,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 23171168,
      "step": 14400
    },
    {
      "epoch": 0.6521197854184114,
      "grad_norm": 0.9213047027587891,
      "learning_rate": 3.563737526890759e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 23179136,
      "step": 14405
    },
    {
      "epoch": 0.6523461373050544,
      "grad_norm": 0.4914231598377228,
      "learning_rate": 3.562849001734049e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 23187008,
      "step": 14410
    },
    {
      "epoch": 0.6525724891916974,
      "grad_norm": 0.30203139781951904,
      "learning_rate": 3.561960312672667e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 23195232,
      "step": 14415
    },
    {
      "epoch": 0.6527988410783404,
      "grad_norm": 0.6865509748458862,
      "learning_rate": 3.5610714598436596e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 23203296,
      "step": 14420
    },
    {
      "epoch": 0.6530251929649834,
      "grad_norm": 0.7824006080627441,
      "learning_rate": 3.5601824433840986e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 23211296,
      "step": 14425
    },
    {
      "epoch": 0.6532515448516263,
      "grad_norm": 0.5173482298851013,
      "learning_rate": 3.559293263431082e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 23219200,
      "step": 14430
    },
    {
      "epoch": 0.6534778967382693,
      "grad_norm": 0.5703450441360474,
      "learning_rate": 3.558403920121732e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 23227264,
      "step": 14435
    },
    {
      "epoch": 0.6537042486249123,
      "grad_norm": 0.6784334778785706,
      "learning_rate": 3.557514413593197e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 23235680,
      "step": 14440
    },
    {
      "epoch": 0.6539306005115553,
      "grad_norm": 2.096731185913086,
      "learning_rate": 3.55662474398265e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 23243520,
      "step": 14445
    },
    {
      "epoch": 0.6541569523981983,
      "grad_norm": 0.4544544219970703,
      "learning_rate": 3.555734911427288e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 23251296,
      "step": 14450
    },
    {
      "epoch": 0.6543833042848413,
      "grad_norm": 0.7903786301612854,
      "learning_rate": 3.5548449160643363e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 23259008,
      "step": 14455
    },
    {
      "epoch": 0.6546096561714841,
      "grad_norm": 0.7887576222419739,
      "learning_rate": 3.553954758031043e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 23267648,
      "step": 14460
    },
    {
      "epoch": 0.6548360080581271,
      "grad_norm": 0.5474879145622253,
      "learning_rate": 3.5530644374646815e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 23276352,
      "step": 14465
    },
    {
      "epoch": 0.6550623599447701,
      "grad_norm": 0.7150168418884277,
      "learning_rate": 3.552173954502549e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 23284736,
      "step": 14470
    },
    {
      "epoch": 0.6552887118314131,
      "grad_norm": 0.5031734704971313,
      "learning_rate": 3.55128330928197e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 23292448,
      "step": 14475
    },
    {
      "epoch": 0.6555150637180561,
      "grad_norm": 0.27026206254959106,
      "learning_rate": 3.550392501940294e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 23300320,
      "step": 14480
    },
    {
      "epoch": 0.6557414156046991,
      "grad_norm": 0.790117084980011,
      "learning_rate": 3.5495015326148945e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 23308352,
      "step": 14485
    },
    {
      "epoch": 0.655967767491342,
      "grad_norm": 0.7012379765510559,
      "learning_rate": 3.548610401443169e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 23316768,
      "step": 14490
    },
    {
      "epoch": 0.656194119377985,
      "grad_norm": 0.660308837890625,
      "learning_rate": 3.547719108562543e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 23325248,
      "step": 14495
    },
    {
      "epoch": 0.656420471264628,
      "grad_norm": 0.6294079422950745,
      "learning_rate": 3.546827654110464e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 23333216,
      "step": 14500
    },
    {
      "epoch": 0.656646823151271,
      "grad_norm": 0.7707546353340149,
      "learning_rate": 3.545936038224405e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 23340672,
      "step": 14505
    },
    {
      "epoch": 0.656873175037914,
      "grad_norm": 0.5043679475784302,
      "learning_rate": 3.545044261041864e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 23348736,
      "step": 14510
    },
    {
      "epoch": 0.6570995269245569,
      "grad_norm": 0.5256573557853699,
      "learning_rate": 3.5441523227003657e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 23357216,
      "step": 14515
    },
    {
      "epoch": 0.6573258788111999,
      "grad_norm": 0.8692076802253723,
      "learning_rate": 3.543260223337459e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 23364928,
      "step": 14520
    },
    {
      "epoch": 0.6575522306978429,
      "grad_norm": 0.4810981750488281,
      "learning_rate": 3.542367963090714e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 23372864,
      "step": 14525
    },
    {
      "epoch": 0.6577785825844858,
      "grad_norm": 0.5513935089111328,
      "learning_rate": 3.5414755420977295e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 23381952,
      "step": 14530
    },
    {
      "epoch": 0.6580049344711288,
      "grad_norm": 0.5218486785888672,
      "learning_rate": 3.54058296049613e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 23389600,
      "step": 14535
    },
    {
      "epoch": 0.6582312863577718,
      "grad_norm": 0.6303165555000305,
      "learning_rate": 3.53969021842356e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 23397472,
      "step": 14540
    },
    {
      "epoch": 0.6584576382444147,
      "grad_norm": 0.9389448761940002,
      "learning_rate": 3.5387973160176926e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 23405152,
      "step": 14545
    },
    {
      "epoch": 0.6586839901310577,
      "grad_norm": 0.5046928524971008,
      "learning_rate": 3.537904253416224e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 23412768,
      "step": 14550
    },
    {
      "epoch": 0.6589103420177007,
      "grad_norm": 0.8530815243721008,
      "learning_rate": 3.537011030756878e-05,
      "loss": 0.1834,
      "num_input_tokens_seen": 23420864,
      "step": 14555
    },
    {
      "epoch": 0.6591366939043437,
      "grad_norm": 0.4016847610473633,
      "learning_rate": 3.536117648177399e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 23428960,
      "step": 14560
    },
    {
      "epoch": 0.6593630457909867,
      "grad_norm": 0.46530768275260925,
      "learning_rate": 3.535224105815558e-05,
      "loss": 0.1941,
      "num_input_tokens_seen": 23437440,
      "step": 14565
    },
    {
      "epoch": 0.6595893976776297,
      "grad_norm": 0.3588670790195465,
      "learning_rate": 3.5343304038091494e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 23445024,
      "step": 14570
    },
    {
      "epoch": 0.6598157495642726,
      "grad_norm": 0.8106832504272461,
      "learning_rate": 3.5334365422959955e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 23453056,
      "step": 14575
    },
    {
      "epoch": 0.6600421014509156,
      "grad_norm": 0.38834965229034424,
      "learning_rate": 3.5325425214139396e-05,
      "loss": 0.1185,
      "num_input_tokens_seen": 23461472,
      "step": 14580
    },
    {
      "epoch": 0.6602684533375586,
      "grad_norm": 0.8253448605537415,
      "learning_rate": 3.531648341300851e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 23469408,
      "step": 14585
    },
    {
      "epoch": 0.6604948052242016,
      "grad_norm": 0.33251717686653137,
      "learning_rate": 3.530754002094623e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 23477472,
      "step": 14590
    },
    {
      "epoch": 0.6607211571108446,
      "grad_norm": 0.6529770493507385,
      "learning_rate": 3.529859503933175e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 23485408,
      "step": 14595
    },
    {
      "epoch": 0.6609475089974874,
      "grad_norm": 0.8990203738212585,
      "learning_rate": 3.52896484695445e-05,
      "loss": 0.159,
      "num_input_tokens_seen": 23493152,
      "step": 14600
    },
    {
      "epoch": 0.6609475089974874,
      "eval_loss": 0.15319599211215973,
      "eval_runtime": 404.6251,
      "eval_samples_per_second": 97.055,
      "eval_steps_per_second": 24.264,
      "num_input_tokens_seen": 23493152,
      "step": 14600
    },
    {
      "epoch": 0.6611738608841304,
      "grad_norm": 0.5411087870597839,
      "learning_rate": 3.528070031296414e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 23500800,
      "step": 14605
    },
    {
      "epoch": 0.6614002127707734,
      "grad_norm": 0.9593368768692017,
      "learning_rate": 3.5271750570970605e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 23509152,
      "step": 14610
    },
    {
      "epoch": 0.6616265646574164,
      "grad_norm": 0.4620039761066437,
      "learning_rate": 3.526279924494405e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 23516992,
      "step": 14615
    },
    {
      "epoch": 0.6618529165440594,
      "grad_norm": 0.30709150433540344,
      "learning_rate": 3.5253846336264874e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 23526048,
      "step": 14620
    },
    {
      "epoch": 0.6620792684307024,
      "grad_norm": 0.4557216465473175,
      "learning_rate": 3.5244891846313736e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 23533760,
      "step": 14625
    },
    {
      "epoch": 0.6623056203173453,
      "grad_norm": 0.5576377511024475,
      "learning_rate": 3.5235935776471527e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 23541344,
      "step": 14630
    },
    {
      "epoch": 0.6625319722039883,
      "grad_norm": 0.4845236539840698,
      "learning_rate": 3.522697812811939e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 23549120,
      "step": 14635
    },
    {
      "epoch": 0.6627583240906313,
      "grad_norm": 0.8772628903388977,
      "learning_rate": 3.521801890263871e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 23557408,
      "step": 14640
    },
    {
      "epoch": 0.6629846759772743,
      "grad_norm": 0.3568573594093323,
      "learning_rate": 3.5209058101411114e-05,
      "loss": 0.2093,
      "num_input_tokens_seen": 23564864,
      "step": 14645
    },
    {
      "epoch": 0.6632110278639173,
      "grad_norm": 0.26620176434516907,
      "learning_rate": 3.520009572581845e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 23573440,
      "step": 14650
    },
    {
      "epoch": 0.6634373797505603,
      "grad_norm": 0.5895771980285645,
      "learning_rate": 3.519113177724285e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 23581184,
      "step": 14655
    },
    {
      "epoch": 0.6636637316372032,
      "grad_norm": 0.3884633183479309,
      "learning_rate": 3.5182166257066656e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 23588864,
      "step": 14660
    },
    {
      "epoch": 0.6638900835238462,
      "grad_norm": 1.4546513557434082,
      "learning_rate": 3.517319916667247e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 23596992,
      "step": 14665
    },
    {
      "epoch": 0.6641164354104891,
      "grad_norm": 1.4179576635360718,
      "learning_rate": 3.516423050744313e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 23605376,
      "step": 14670
    },
    {
      "epoch": 0.6643427872971321,
      "grad_norm": 0.956499457359314,
      "learning_rate": 3.5155260280761704e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 23613760,
      "step": 14675
    },
    {
      "epoch": 0.6645691391837751,
      "grad_norm": 0.7411190867424011,
      "learning_rate": 3.514628848801154e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 23621472,
      "step": 14680
    },
    {
      "epoch": 0.664795491070418,
      "grad_norm": 1.3853198289871216,
      "learning_rate": 3.5137315130576174e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 23629408,
      "step": 14685
    },
    {
      "epoch": 0.665021842957061,
      "grad_norm": 0.7811347246170044,
      "learning_rate": 3.512834020983942e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 23637760,
      "step": 14690
    },
    {
      "epoch": 0.665248194843704,
      "grad_norm": 0.672248125076294,
      "learning_rate": 3.5119363727185334e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 23646432,
      "step": 14695
    },
    {
      "epoch": 0.665474546730347,
      "grad_norm": 0.5856724381446838,
      "learning_rate": 3.511038568399819e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 23654880,
      "step": 14700
    },
    {
      "epoch": 0.66570089861699,
      "grad_norm": 0.40411242842674255,
      "learning_rate": 3.510140608166251e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 23662912,
      "step": 14705
    },
    {
      "epoch": 0.665927250503633,
      "grad_norm": 0.3136584162712097,
      "learning_rate": 3.509242492156308e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 23671232,
      "step": 14710
    },
    {
      "epoch": 0.6661536023902759,
      "grad_norm": 0.5737354159355164,
      "learning_rate": 3.5083442205084896e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 23679104,
      "step": 14715
    },
    {
      "epoch": 0.6663799542769189,
      "grad_norm": 0.3340108394622803,
      "learning_rate": 3.507445793361321e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 23686912,
      "step": 14720
    },
    {
      "epoch": 0.6666063061635619,
      "grad_norm": 0.7541811466217041,
      "learning_rate": 3.5065472108533505e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 23695168,
      "step": 14725
    },
    {
      "epoch": 0.6668326580502049,
      "grad_norm": 1.1916344165802002,
      "learning_rate": 3.5056484731231504e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 23703552,
      "step": 14730
    },
    {
      "epoch": 0.6670590099368479,
      "grad_norm": 0.6221421360969543,
      "learning_rate": 3.504749580309319e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 23711840,
      "step": 14735
    },
    {
      "epoch": 0.6672853618234909,
      "grad_norm": 0.6900644302368164,
      "learning_rate": 3.5038505325504753e-05,
      "loss": 0.1722,
      "num_input_tokens_seen": 23719936,
      "step": 14740
    },
    {
      "epoch": 0.6675117137101337,
      "grad_norm": 0.43083664774894714,
      "learning_rate": 3.502951329985264e-05,
      "loss": 0.159,
      "num_input_tokens_seen": 23727904,
      "step": 14745
    },
    {
      "epoch": 0.6677380655967767,
      "grad_norm": 1.7643659114837646,
      "learning_rate": 3.502051972752354e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 23735712,
      "step": 14750
    },
    {
      "epoch": 0.6679644174834197,
      "grad_norm": 0.6505005359649658,
      "learning_rate": 3.5011524609904374e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 23743904,
      "step": 14755
    },
    {
      "epoch": 0.6681907693700627,
      "grad_norm": 0.7600333094596863,
      "learning_rate": 3.50025279483823e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 23751264,
      "step": 14760
    },
    {
      "epoch": 0.6684171212567057,
      "grad_norm": 0.34926560521125793,
      "learning_rate": 3.499352974434472e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 23759136,
      "step": 14765
    },
    {
      "epoch": 0.6686434731433486,
      "grad_norm": 0.9408482909202576,
      "learning_rate": 3.498452999917926e-05,
      "loss": 0.1177,
      "num_input_tokens_seen": 23766880,
      "step": 14770
    },
    {
      "epoch": 0.6688698250299916,
      "grad_norm": 0.5629829168319702,
      "learning_rate": 3.4975528714273795e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 23774624,
      "step": 14775
    },
    {
      "epoch": 0.6690961769166346,
      "grad_norm": 0.5731796622276306,
      "learning_rate": 3.4966525891016454e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 23782656,
      "step": 14780
    },
    {
      "epoch": 0.6693225288032776,
      "grad_norm": 1.0415328741073608,
      "learning_rate": 3.495752153079557e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 23790944,
      "step": 14785
    },
    {
      "epoch": 0.6695488806899206,
      "grad_norm": 0.4311385452747345,
      "learning_rate": 3.494851563499974e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 23799040,
      "step": 14790
    },
    {
      "epoch": 0.6697752325765636,
      "grad_norm": 0.43589043617248535,
      "learning_rate": 3.493950820501777e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 23807232,
      "step": 14795
    },
    {
      "epoch": 0.6700015844632065,
      "grad_norm": 0.6781229972839355,
      "learning_rate": 3.493049924223872e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 23815360,
      "step": 14800
    },
    {
      "epoch": 0.6700015844632065,
      "eval_loss": 0.15266293287277222,
      "eval_runtime": 405.3322,
      "eval_samples_per_second": 96.886,
      "eval_steps_per_second": 24.222,
      "num_input_tokens_seen": 23815360,
      "step": 14800
    },
    {
      "epoch": 0.6702279363498495,
      "grad_norm": 0.554610013961792,
      "learning_rate": 3.49214887480519e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 23823360,
      "step": 14805
    },
    {
      "epoch": 0.6704542882364924,
      "grad_norm": 0.6580770015716553,
      "learning_rate": 3.4912476723846834e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 23831488,
      "step": 14810
    },
    {
      "epoch": 0.6706806401231354,
      "grad_norm": 0.6128851175308228,
      "learning_rate": 3.490346317101328e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 23838880,
      "step": 14815
    },
    {
      "epoch": 0.6709069920097784,
      "grad_norm": 1.3933302164077759,
      "learning_rate": 3.4894448090941266e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 23846592,
      "step": 14820
    },
    {
      "epoch": 0.6711333438964214,
      "grad_norm": 1.1222189664840698,
      "learning_rate": 3.488543148502101e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 23855072,
      "step": 14825
    },
    {
      "epoch": 0.6713596957830643,
      "grad_norm": 0.7056429982185364,
      "learning_rate": 3.487641335464299e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 23862560,
      "step": 14830
    },
    {
      "epoch": 0.6715860476697073,
      "grad_norm": 0.5643945336341858,
      "learning_rate": 3.4867393701197914e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 23870560,
      "step": 14835
    },
    {
      "epoch": 0.6718123995563503,
      "grad_norm": 0.8977655172348022,
      "learning_rate": 3.485837252607673e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 23878784,
      "step": 14840
    },
    {
      "epoch": 0.6720387514429933,
      "grad_norm": 0.9904186725616455,
      "learning_rate": 3.4849349830670615e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 23886368,
      "step": 14845
    },
    {
      "epoch": 0.6722651033296363,
      "grad_norm": 0.4535013437271118,
      "learning_rate": 3.4840325616370976e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 23894272,
      "step": 14850
    },
    {
      "epoch": 0.6724914552162792,
      "grad_norm": 0.8757176399230957,
      "learning_rate": 3.483129988456947e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 23902464,
      "step": 14855
    },
    {
      "epoch": 0.6727178071029222,
      "grad_norm": 0.821678102016449,
      "learning_rate": 3.482227263665797e-05,
      "loss": 0.21,
      "num_input_tokens_seen": 23910496,
      "step": 14860
    },
    {
      "epoch": 0.6729441589895652,
      "grad_norm": 0.48952916264533997,
      "learning_rate": 3.48132438740286e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 23918624,
      "step": 14865
    },
    {
      "epoch": 0.6731705108762082,
      "grad_norm": 1.0260671377182007,
      "learning_rate": 3.48042135980737e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 23927168,
      "step": 14870
    },
    {
      "epoch": 0.6733968627628512,
      "grad_norm": 1.075318455696106,
      "learning_rate": 3.479518181018586e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 23935744,
      "step": 14875
    },
    {
      "epoch": 0.6736232146494942,
      "grad_norm": 0.29853326082229614,
      "learning_rate": 3.4786148511757886e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 23943584,
      "step": 14880
    },
    {
      "epoch": 0.673849566536137,
      "grad_norm": 0.3390541970729828,
      "learning_rate": 3.477711370418284e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 23951776,
      "step": 14885
    },
    {
      "epoch": 0.67407591842278,
      "grad_norm": 0.9628533124923706,
      "learning_rate": 3.476807738885399e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 23960256,
      "step": 14890
    },
    {
      "epoch": 0.674302270309423,
      "grad_norm": 0.7976846694946289,
      "learning_rate": 3.475903956716485e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 23968192,
      "step": 14895
    },
    {
      "epoch": 0.674528622196066,
      "grad_norm": 0.7594956755638123,
      "learning_rate": 3.475000024050917e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 23976384,
      "step": 14900
    },
    {
      "epoch": 0.674754974082709,
      "grad_norm": 1.6262171268463135,
      "learning_rate": 3.4740959410280926e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 23984160,
      "step": 14905
    },
    {
      "epoch": 0.674981325969352,
      "grad_norm": 0.6004962921142578,
      "learning_rate": 3.4731917077874324e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 23992576,
      "step": 14910
    },
    {
      "epoch": 0.6752076778559949,
      "grad_norm": 0.732342541217804,
      "learning_rate": 3.4722873244683816e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 24000192,
      "step": 14915
    },
    {
      "epoch": 0.6754340297426379,
      "grad_norm": 0.5387979745864868,
      "learning_rate": 3.4713827912104065e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 24008064,
      "step": 14920
    },
    {
      "epoch": 0.6756603816292809,
      "grad_norm": 0.4743080735206604,
      "learning_rate": 3.470478108152998e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 24016544,
      "step": 14925
    },
    {
      "epoch": 0.6758867335159239,
      "grad_norm": 0.7692604064941406,
      "learning_rate": 3.4695732754356695e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 24025024,
      "step": 14930
    },
    {
      "epoch": 0.6761130854025669,
      "grad_norm": 0.8323035836219788,
      "learning_rate": 3.4686682931979576e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 24033728,
      "step": 14935
    },
    {
      "epoch": 0.6763394372892098,
      "grad_norm": 0.4457862675189972,
      "learning_rate": 3.467763161579422e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 24041536,
      "step": 14940
    },
    {
      "epoch": 0.6765657891758527,
      "grad_norm": 0.420393705368042,
      "learning_rate": 3.466857880719645e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 24049504,
      "step": 14945
    },
    {
      "epoch": 0.6767921410624957,
      "grad_norm": 0.5569444298744202,
      "learning_rate": 3.465952450758233e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 24057696,
      "step": 14950
    },
    {
      "epoch": 0.6770184929491387,
      "grad_norm": 0.4172670543193817,
      "learning_rate": 3.4650468718348126e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 24065728,
      "step": 14955
    },
    {
      "epoch": 0.6772448448357817,
      "grad_norm": 1.103678584098816,
      "learning_rate": 3.464141144089038e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 24073440,
      "step": 14960
    },
    {
      "epoch": 0.6774711967224247,
      "grad_norm": 0.5666486024856567,
      "learning_rate": 3.463235267660583e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 24081088,
      "step": 14965
    },
    {
      "epoch": 0.6776975486090676,
      "grad_norm": 0.7099390029907227,
      "learning_rate": 3.462329242689145e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 24088896,
      "step": 14970
    },
    {
      "epoch": 0.6779239004957106,
      "grad_norm": 0.5031909942626953,
      "learning_rate": 3.461423069314444e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 24096992,
      "step": 14975
    },
    {
      "epoch": 0.6781502523823536,
      "grad_norm": 0.5433902144432068,
      "learning_rate": 3.460516747676224e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 24104608,
      "step": 14980
    },
    {
      "epoch": 0.6783766042689966,
      "grad_norm": 0.6879739761352539,
      "learning_rate": 3.459610277914251e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 24112480,
      "step": 14985
    },
    {
      "epoch": 0.6786029561556396,
      "grad_norm": 0.7405552864074707,
      "learning_rate": 3.458703660168314e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 24121184,
      "step": 14990
    },
    {
      "epoch": 0.6788293080422826,
      "grad_norm": 0.7975612878799438,
      "learning_rate": 3.457796894578224e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 24129728,
      "step": 14995
    },
    {
      "epoch": 0.6790556599289255,
      "grad_norm": 0.8893002867698669,
      "learning_rate": 3.456889981283817e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 24137696,
      "step": 15000
    },
    {
      "epoch": 0.6790556599289255,
      "eval_loss": 0.15154898166656494,
      "eval_runtime": 404.1335,
      "eval_samples_per_second": 97.173,
      "eval_steps_per_second": 24.294,
      "num_input_tokens_seen": 24137696,
      "step": 15000
    },
    {
      "epoch": 0.6792820118155685,
      "grad_norm": 0.7040050625801086,
      "learning_rate": 3.45598292042495e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 24145696,
      "step": 15005
    },
    {
      "epoch": 0.6795083637022115,
      "grad_norm": 0.5646432638168335,
      "learning_rate": 3.4550757121415035e-05,
      "loss": 0.176,
      "num_input_tokens_seen": 24153824,
      "step": 15010
    },
    {
      "epoch": 0.6797347155888545,
      "grad_norm": 0.4420105218887329,
      "learning_rate": 3.454168356573378e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 24161600,
      "step": 15015
    },
    {
      "epoch": 0.6799610674754974,
      "grad_norm": 0.939793586730957,
      "learning_rate": 3.453260853860503e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 24169440,
      "step": 15020
    },
    {
      "epoch": 0.6801874193621403,
      "grad_norm": 0.5065263509750366,
      "learning_rate": 3.452353204142824e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 24177056,
      "step": 15025
    },
    {
      "epoch": 0.6804137712487833,
      "grad_norm": 1.0985723733901978,
      "learning_rate": 3.4514454075603136e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 24184928,
      "step": 15030
    },
    {
      "epoch": 0.6806401231354263,
      "grad_norm": 0.4782111942768097,
      "learning_rate": 3.450537464252964e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 24193344,
      "step": 15035
    },
    {
      "epoch": 0.6808664750220693,
      "grad_norm": 0.40116679668426514,
      "learning_rate": 3.4496293743607925e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 24201248,
      "step": 15040
    },
    {
      "epoch": 0.6810928269087123,
      "grad_norm": 0.7593627572059631,
      "learning_rate": 3.448721138023838e-05,
      "loss": 0.111,
      "num_input_tokens_seen": 24209024,
      "step": 15045
    },
    {
      "epoch": 0.6813191787953553,
      "grad_norm": 1.091964840888977,
      "learning_rate": 3.447812755382162e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 24217088,
      "step": 15050
    },
    {
      "epoch": 0.6815455306819982,
      "grad_norm": 0.5896603465080261,
      "learning_rate": 3.446904226575847e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 24224992,
      "step": 15055
    },
    {
      "epoch": 0.6817718825686412,
      "grad_norm": 0.9798548817634583,
      "learning_rate": 3.445995551745002e-05,
      "loss": 0.1845,
      "num_input_tokens_seen": 24232896,
      "step": 15060
    },
    {
      "epoch": 0.6819982344552842,
      "grad_norm": 0.44167473912239075,
      "learning_rate": 3.445086731029753e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 24241216,
      "step": 15065
    },
    {
      "epoch": 0.6822245863419272,
      "grad_norm": 0.9220307469367981,
      "learning_rate": 3.444177764570255e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 24249024,
      "step": 15070
    },
    {
      "epoch": 0.6824509382285702,
      "grad_norm": 0.37694069743156433,
      "learning_rate": 3.44326865250668e-05,
      "loss": 0.181,
      "num_input_tokens_seen": 24256736,
      "step": 15075
    },
    {
      "epoch": 0.6826772901152132,
      "grad_norm": 0.8010609149932861,
      "learning_rate": 3.442359394979225e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 24264928,
      "step": 15080
    },
    {
      "epoch": 0.682903642001856,
      "grad_norm": 0.7699037194252014,
      "learning_rate": 3.441449992128108e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 24273152,
      "step": 15085
    },
    {
      "epoch": 0.683129993888499,
      "grad_norm": 0.8180553317070007,
      "learning_rate": 3.440540444093573e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 24281152,
      "step": 15090
    },
    {
      "epoch": 0.683356345775142,
      "grad_norm": 0.4953412115573883,
      "learning_rate": 3.43963075101588e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 24288896,
      "step": 15095
    },
    {
      "epoch": 0.683582697661785,
      "grad_norm": 0.6676454544067383,
      "learning_rate": 3.438720913035318e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 24296672,
      "step": 15100
    },
    {
      "epoch": 0.683809049548428,
      "grad_norm": 0.3611292839050293,
      "learning_rate": 3.437810930292195e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 24304928,
      "step": 15105
    },
    {
      "epoch": 0.6840354014350709,
      "grad_norm": 0.48353952169418335,
      "learning_rate": 3.43690080292684e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 24312832,
      "step": 15110
    },
    {
      "epoch": 0.6842617533217139,
      "grad_norm": 1.160891056060791,
      "learning_rate": 3.435990531079608e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 24320352,
      "step": 15115
    },
    {
      "epoch": 0.6844881052083569,
      "grad_norm": 0.46196553111076355,
      "learning_rate": 3.435080114890874e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 24327808,
      "step": 15120
    },
    {
      "epoch": 0.6847144570949999,
      "grad_norm": 0.973038375377655,
      "learning_rate": 3.434169554501035e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 24335808,
      "step": 15125
    },
    {
      "epoch": 0.6849408089816429,
      "grad_norm": 0.37795257568359375,
      "learning_rate": 3.433258850050511e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 24343744,
      "step": 15130
    },
    {
      "epoch": 0.6851671608682859,
      "grad_norm": 0.6123005747795105,
      "learning_rate": 3.4323480016797446e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 24351488,
      "step": 15135
    },
    {
      "epoch": 0.6853935127549288,
      "grad_norm": 0.4131680130958557,
      "learning_rate": 3.4314370095291995e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 24359552,
      "step": 15140
    },
    {
      "epoch": 0.6856198646415718,
      "grad_norm": 0.8312063813209534,
      "learning_rate": 3.430525873739363e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 24367744,
      "step": 15145
    },
    {
      "epoch": 0.6858462165282148,
      "grad_norm": 0.44290891289711,
      "learning_rate": 3.429614594450743e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 24375680,
      "step": 15150
    },
    {
      "epoch": 0.6860725684148578,
      "grad_norm": 0.34957313537597656,
      "learning_rate": 3.428703171803869e-05,
      "loss": 0.1087,
      "num_input_tokens_seen": 24383552,
      "step": 15155
    },
    {
      "epoch": 0.6862989203015007,
      "grad_norm": 0.4706166088581085,
      "learning_rate": 3.4277916059392964e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 24391168,
      "step": 15160
    },
    {
      "epoch": 0.6865252721881437,
      "grad_norm": 0.6181322932243347,
      "learning_rate": 3.426879896997598e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 24398880,
      "step": 15165
    },
    {
      "epoch": 0.6867516240747866,
      "grad_norm": 0.6000714898109436,
      "learning_rate": 3.425968045119372e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 24406464,
      "step": 15170
    },
    {
      "epoch": 0.6869779759614296,
      "grad_norm": 1.1469807624816895,
      "learning_rate": 3.425056050445237e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 24414592,
      "step": 15175
    },
    {
      "epoch": 0.6872043278480726,
      "grad_norm": 0.9114106297492981,
      "learning_rate": 3.4241439131158336e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 24422848,
      "step": 15180
    },
    {
      "epoch": 0.6874306797347156,
      "grad_norm": 1.0455831289291382,
      "learning_rate": 3.423231633271825e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 24431072,
      "step": 15185
    },
    {
      "epoch": 0.6876570316213586,
      "grad_norm": 0.5982682108879089,
      "learning_rate": 3.4223192110538985e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 24438880,
      "step": 15190
    },
    {
      "epoch": 0.6878833835080015,
      "grad_norm": 0.8535938858985901,
      "learning_rate": 3.4214066466027575e-05,
      "loss": 0.1814,
      "num_input_tokens_seen": 24447712,
      "step": 15195
    },
    {
      "epoch": 0.6881097353946445,
      "grad_norm": 0.5524920225143433,
      "learning_rate": 3.4204939400591325e-05,
      "loss": 0.1118,
      "num_input_tokens_seen": 24455584,
      "step": 15200
    },
    {
      "epoch": 0.6881097353946445,
      "eval_loss": 0.15167178213596344,
      "eval_runtime": 404.9074,
      "eval_samples_per_second": 96.988,
      "eval_steps_per_second": 24.248,
      "num_input_tokens_seen": 24455584,
      "step": 15200
    },
    {
      "epoch": 0.6883360872812875,
      "grad_norm": 0.7145678997039795,
      "learning_rate": 3.419581091563775e-05,
      "loss": 0.1893,
      "num_input_tokens_seen": 24463712,
      "step": 15205
    },
    {
      "epoch": 0.6885624391679305,
      "grad_norm": 0.8776836395263672,
      "learning_rate": 3.418668101257456e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 24472672,
      "step": 15210
    },
    {
      "epoch": 0.6887887910545735,
      "grad_norm": 1.0482558012008667,
      "learning_rate": 3.417754969280971e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 24480416,
      "step": 15215
    },
    {
      "epoch": 0.6890151429412165,
      "grad_norm": 0.5828083157539368,
      "learning_rate": 3.416841695775137e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 24488800,
      "step": 15220
    },
    {
      "epoch": 0.6892414948278593,
      "grad_norm": 0.6094642281532288,
      "learning_rate": 3.415928280880792e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 24496736,
      "step": 15225
    },
    {
      "epoch": 0.6894678467145023,
      "grad_norm": 0.8641563653945923,
      "learning_rate": 3.4150147247387965e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 24504640,
      "step": 15230
    },
    {
      "epoch": 0.6896941986011453,
      "grad_norm": 0.34176769852638245,
      "learning_rate": 3.4141010274900306e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 24512864,
      "step": 15235
    },
    {
      "epoch": 0.6899205504877883,
      "grad_norm": 0.9371000528335571,
      "learning_rate": 3.413187189275399e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 24520576,
      "step": 15240
    },
    {
      "epoch": 0.6901469023744313,
      "grad_norm": 0.5043493509292603,
      "learning_rate": 3.4122732102358265e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 24528480,
      "step": 15245
    },
    {
      "epoch": 0.6903732542610743,
      "grad_norm": 0.43685972690582275,
      "learning_rate": 3.411359090512261e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 24536288,
      "step": 15250
    },
    {
      "epoch": 0.6905996061477172,
      "grad_norm": 1.1485559940338135,
      "learning_rate": 3.410444830245672e-05,
      "loss": 0.1859,
      "num_input_tokens_seen": 24544192,
      "step": 15255
    },
    {
      "epoch": 0.6908259580343602,
      "grad_norm": 0.6600596308708191,
      "learning_rate": 3.409530429577048e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 24552256,
      "step": 15260
    },
    {
      "epoch": 0.6910523099210032,
      "grad_norm": 0.3695008158683777,
      "learning_rate": 3.408615888647402e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 24560224,
      "step": 15265
    },
    {
      "epoch": 0.6912786618076462,
      "grad_norm": 0.3665302097797394,
      "learning_rate": 3.4077012075977675e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 24568160,
      "step": 15270
    },
    {
      "epoch": 0.6915050136942892,
      "grad_norm": 0.98244309425354,
      "learning_rate": 3.4067863865692e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 24576576,
      "step": 15275
    },
    {
      "epoch": 0.6917313655809321,
      "grad_norm": 0.7614085674285889,
      "learning_rate": 3.4058714257027755e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 24584480,
      "step": 15280
    },
    {
      "epoch": 0.6919577174675751,
      "grad_norm": 0.421002060174942,
      "learning_rate": 3.404956325139594e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 24592032,
      "step": 15285
    },
    {
      "epoch": 0.6921840693542181,
      "grad_norm": 0.890084981918335,
      "learning_rate": 3.404041085020775e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 24601376,
      "step": 15290
    },
    {
      "epoch": 0.692410421240861,
      "grad_norm": 0.37915924191474915,
      "learning_rate": 3.403125705487459e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 24609536,
      "step": 15295
    },
    {
      "epoch": 0.692636773127504,
      "grad_norm": 0.538015604019165,
      "learning_rate": 3.402210186680811e-05,
      "loss": 0.1553,
      "num_input_tokens_seen": 24617728,
      "step": 15300
    },
    {
      "epoch": 0.692863125014147,
      "grad_norm": 0.5074330568313599,
      "learning_rate": 3.4012945287420137e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 24625312,
      "step": 15305
    },
    {
      "epoch": 0.6930894769007899,
      "grad_norm": 0.6405069828033447,
      "learning_rate": 3.400378731812274e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 24633632,
      "step": 15310
    },
    {
      "epoch": 0.6933158287874329,
      "grad_norm": 1.314967393875122,
      "learning_rate": 3.399462796032817e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 24641216,
      "step": 15315
    },
    {
      "epoch": 0.6935421806740759,
      "grad_norm": 0.5642899870872498,
      "learning_rate": 3.3985467215448954e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 24648992,
      "step": 15320
    },
    {
      "epoch": 0.6937685325607189,
      "grad_norm": 0.5407913327217102,
      "learning_rate": 3.3976305084897776e-05,
      "loss": 0.1822,
      "num_input_tokens_seen": 24657184,
      "step": 15325
    },
    {
      "epoch": 0.6939948844473619,
      "grad_norm": 0.8211143016815186,
      "learning_rate": 3.3967141570087544e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 24664928,
      "step": 15330
    },
    {
      "epoch": 0.6942212363340049,
      "grad_norm": 0.4286591112613678,
      "learning_rate": 3.39579766724314e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 24673088,
      "step": 15335
    },
    {
      "epoch": 0.6944475882206478,
      "grad_norm": 0.8613908290863037,
      "learning_rate": 3.3948810393342677e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 24681440,
      "step": 15340
    },
    {
      "epoch": 0.6946739401072908,
      "grad_norm": 1.3334870338439941,
      "learning_rate": 3.3939642734234936e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 24689984,
      "step": 15345
    },
    {
      "epoch": 0.6949002919939338,
      "grad_norm": 0.5459897518157959,
      "learning_rate": 3.393047369652194e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 24697760,
      "step": 15350
    },
    {
      "epoch": 0.6951266438805768,
      "grad_norm": 1.1721446514129639,
      "learning_rate": 3.3921303281617664e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 24705984,
      "step": 15355
    },
    {
      "epoch": 0.6953529957672198,
      "grad_norm": 0.3980293869972229,
      "learning_rate": 3.391213149093632e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 24714336,
      "step": 15360
    },
    {
      "epoch": 0.6955793476538626,
      "grad_norm": 0.4833526611328125,
      "learning_rate": 3.3902958325892303e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 24722400,
      "step": 15365
    },
    {
      "epoch": 0.6958056995405056,
      "grad_norm": 0.3448801040649414,
      "learning_rate": 3.389378378790023e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 24730560,
      "step": 15370
    },
    {
      "epoch": 0.6960320514271486,
      "grad_norm": 0.3722841441631317,
      "learning_rate": 3.388460787837493e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 24738400,
      "step": 15375
    },
    {
      "epoch": 0.6962584033137916,
      "grad_norm": 0.415618896484375,
      "learning_rate": 3.387543059873145e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 24746528,
      "step": 15380
    },
    {
      "epoch": 0.6964847552004346,
      "grad_norm": 0.4322506785392761,
      "learning_rate": 3.386625195038503e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 24754880,
      "step": 15385
    },
    {
      "epoch": 0.6967111070870776,
      "grad_norm": 0.3667221665382385,
      "learning_rate": 3.3857071934751136e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 24763104,
      "step": 15390
    },
    {
      "epoch": 0.6969374589737205,
      "grad_norm": 0.772965669631958,
      "learning_rate": 3.384789055324544e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 24771232,
      "step": 15395
    },
    {
      "epoch": 0.6971638108603635,
      "grad_norm": 0.49937793612480164,
      "learning_rate": 3.3838707807283843e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 24779040,
      "step": 15400
    },
    {
      "epoch": 0.6971638108603635,
      "eval_loss": 0.15125752985477448,
      "eval_runtime": 404.5222,
      "eval_samples_per_second": 97.08,
      "eval_steps_per_second": 24.271,
      "num_input_tokens_seen": 24779040,
      "step": 15400
    },
    {
      "epoch": 0.6973901627470065,
      "grad_norm": 0.5069116950035095,
      "learning_rate": 3.382952369828243e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 24786784,
      "step": 15405
    },
    {
      "epoch": 0.6976165146336495,
      "grad_norm": 1.4463765621185303,
      "learning_rate": 3.38203382276575e-05,
      "loss": 0.1843,
      "num_input_tokens_seen": 24795008,
      "step": 15410
    },
    {
      "epoch": 0.6978428665202925,
      "grad_norm": 0.7449397444725037,
      "learning_rate": 3.381115139682557e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 24802976,
      "step": 15415
    },
    {
      "epoch": 0.6980692184069355,
      "grad_norm": 0.6669427156448364,
      "learning_rate": 3.3801963207203366e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 24811200,
      "step": 15420
    },
    {
      "epoch": 0.6982955702935784,
      "grad_norm": 0.6467053890228271,
      "learning_rate": 3.379277366020782e-05,
      "loss": 0.1735,
      "num_input_tokens_seen": 24819520,
      "step": 15425
    },
    {
      "epoch": 0.6985219221802214,
      "grad_norm": 0.3840874433517456,
      "learning_rate": 3.3783582757256085e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 24827488,
      "step": 15430
    },
    {
      "epoch": 0.6987482740668643,
      "grad_norm": 1.488663911819458,
      "learning_rate": 3.3774390499765504e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 24835200,
      "step": 15435
    },
    {
      "epoch": 0.6989746259535073,
      "grad_norm": 0.36377131938934326,
      "learning_rate": 3.376519688915364e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 24843744,
      "step": 15440
    },
    {
      "epoch": 0.6992009778401503,
      "grad_norm": 0.5324881076812744,
      "learning_rate": 3.3756001926838273e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 24851584,
      "step": 15445
    },
    {
      "epoch": 0.6994273297267932,
      "grad_norm": 0.6439149379730225,
      "learning_rate": 3.374680561423737e-05,
      "loss": 0.183,
      "num_input_tokens_seen": 24859904,
      "step": 15450
    },
    {
      "epoch": 0.6996536816134362,
      "grad_norm": 0.80563884973526,
      "learning_rate": 3.373760795276912e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 24867808,
      "step": 15455
    },
    {
      "epoch": 0.6998800335000792,
      "grad_norm": 0.9565438628196716,
      "learning_rate": 3.372840894385192e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 24875360,
      "step": 15460
    },
    {
      "epoch": 0.7001063853867222,
      "grad_norm": 0.7580255270004272,
      "learning_rate": 3.3719208588904375e-05,
      "loss": 0.1752,
      "num_input_tokens_seen": 24883136,
      "step": 15465
    },
    {
      "epoch": 0.7003327372733652,
      "grad_norm": 0.5249824523925781,
      "learning_rate": 3.371000688934529e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 24891168,
      "step": 15470
    },
    {
      "epoch": 0.7005590891600082,
      "grad_norm": 0.35353758931159973,
      "learning_rate": 3.370080384659369e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 24899424,
      "step": 15475
    },
    {
      "epoch": 0.7007854410466511,
      "grad_norm": 0.5294599533081055,
      "learning_rate": 3.36915994620688e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 24907232,
      "step": 15480
    },
    {
      "epoch": 0.7010117929332941,
      "grad_norm": 0.5435915589332581,
      "learning_rate": 3.3682393737190035e-05,
      "loss": 0.1614,
      "num_input_tokens_seen": 24915552,
      "step": 15485
    },
    {
      "epoch": 0.7012381448199371,
      "grad_norm": 0.6884921193122864,
      "learning_rate": 3.3673186673377054e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 24923520,
      "step": 15490
    },
    {
      "epoch": 0.7014644967065801,
      "grad_norm": 0.5022943019866943,
      "learning_rate": 3.366397827204969e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 24931456,
      "step": 15495
    },
    {
      "epoch": 0.7016908485932231,
      "grad_norm": 0.5716585516929626,
      "learning_rate": 3.3654768534628e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 24939776,
      "step": 15500
    },
    {
      "epoch": 0.701917200479866,
      "grad_norm": 0.3148665726184845,
      "learning_rate": 3.3645557462532245e-05,
      "loss": 0.1005,
      "num_input_tokens_seen": 24947648,
      "step": 15505
    },
    {
      "epoch": 0.7021435523665089,
      "grad_norm": 0.44138288497924805,
      "learning_rate": 3.363634505718288e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 24955200,
      "step": 15510
    },
    {
      "epoch": 0.7023699042531519,
      "grad_norm": 0.8351033926010132,
      "learning_rate": 3.362713132000057e-05,
      "loss": 0.1883,
      "num_input_tokens_seen": 24962976,
      "step": 15515
    },
    {
      "epoch": 0.7025962561397949,
      "grad_norm": 0.4397302269935608,
      "learning_rate": 3.36179162524062e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 24970688,
      "step": 15520
    },
    {
      "epoch": 0.7028226080264379,
      "grad_norm": 0.9081175327301025,
      "learning_rate": 3.3608699855820846e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 24979008,
      "step": 15525
    },
    {
      "epoch": 0.7030489599130809,
      "grad_norm": 0.7592847347259521,
      "learning_rate": 3.359948213166578e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 24987424,
      "step": 15530
    },
    {
      "epoch": 0.7032753117997238,
      "grad_norm": 0.9399872422218323,
      "learning_rate": 3.359026308136252e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 24995008,
      "step": 15535
    },
    {
      "epoch": 0.7035016636863668,
      "grad_norm": 1.1316555738449097,
      "learning_rate": 3.358104270633272e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 25002816,
      "step": 15540
    },
    {
      "epoch": 0.7037280155730098,
      "grad_norm": 0.3328101933002472,
      "learning_rate": 3.357182100799831e-05,
      "loss": 0.115,
      "num_input_tokens_seen": 25011488,
      "step": 15545
    },
    {
      "epoch": 0.7039543674596528,
      "grad_norm": 0.5864851474761963,
      "learning_rate": 3.3562597987781384e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 25019712,
      "step": 15550
    },
    {
      "epoch": 0.7041807193462958,
      "grad_norm": 0.2371857464313507,
      "learning_rate": 3.355337364710424e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 25027936,
      "step": 15555
    },
    {
      "epoch": 0.7044070712329388,
      "grad_norm": 0.584632933139801,
      "learning_rate": 3.354414798738939e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 25035584,
      "step": 15560
    },
    {
      "epoch": 0.7046334231195817,
      "grad_norm": 0.8806820511817932,
      "learning_rate": 3.353492101005955e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 25043424,
      "step": 15565
    },
    {
      "epoch": 0.7048597750062247,
      "grad_norm": 0.5598389506340027,
      "learning_rate": 3.352569271653763e-05,
      "loss": 0.1116,
      "num_input_tokens_seen": 25052480,
      "step": 15570
    },
    {
      "epoch": 0.7050861268928676,
      "grad_norm": 0.2531450390815735,
      "learning_rate": 3.351646310824675e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 25059840,
      "step": 15575
    },
    {
      "epoch": 0.7053124787795106,
      "grad_norm": 0.5271688103675842,
      "learning_rate": 3.350723218661023e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 25067680,
      "step": 15580
    },
    {
      "epoch": 0.7055388306661536,
      "grad_norm": 0.5504834651947021,
      "learning_rate": 3.349799995305162e-05,
      "loss": 0.1699,
      "num_input_tokens_seen": 25075904,
      "step": 15585
    },
    {
      "epoch": 0.7057651825527966,
      "grad_norm": 0.3268676996231079,
      "learning_rate": 3.348876640899461e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 25083904,
      "step": 15590
    },
    {
      "epoch": 0.7059915344394395,
      "grad_norm": 0.5085314512252808,
      "learning_rate": 3.3479531555863144e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 25091712,
      "step": 15595
    },
    {
      "epoch": 0.7062178863260825,
      "grad_norm": 0.5751253962516785,
      "learning_rate": 3.3470295395081344e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 25099584,
      "step": 15600
    },
    {
      "epoch": 0.7062178863260825,
      "eval_loss": 0.15077322721481323,
      "eval_runtime": 405.041,
      "eval_samples_per_second": 96.956,
      "eval_steps_per_second": 24.24,
      "num_input_tokens_seen": 25099584,
      "step": 15600
    },
    {
      "epoch": 0.7064442382127255,
      "grad_norm": 0.7349709272384644,
      "learning_rate": 3.3461057928073556e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 25107040,
      "step": 15605
    },
    {
      "epoch": 0.7066705900993685,
      "grad_norm": 0.5594084858894348,
      "learning_rate": 3.345181915626431e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 25114848,
      "step": 15610
    },
    {
      "epoch": 0.7068969419860115,
      "grad_norm": 0.39252015948295593,
      "learning_rate": 3.344257908107834e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 25122912,
      "step": 15615
    },
    {
      "epoch": 0.7071232938726544,
      "grad_norm": 0.4237061142921448,
      "learning_rate": 3.343333770394058e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 25132128,
      "step": 15620
    },
    {
      "epoch": 0.7073496457592974,
      "grad_norm": 0.48546311259269714,
      "learning_rate": 3.342409502627616e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 25139680,
      "step": 15625
    },
    {
      "epoch": 0.7075759976459404,
      "grad_norm": 0.5963193774223328,
      "learning_rate": 3.341485104951043e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 25147584,
      "step": 15630
    },
    {
      "epoch": 0.7078023495325834,
      "grad_norm": 0.6249524354934692,
      "learning_rate": 3.340560577506892e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 25155040,
      "step": 15635
    },
    {
      "epoch": 0.7080287014192264,
      "grad_norm": 0.563754677772522,
      "learning_rate": 3.339635920437735e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 25163168,
      "step": 15640
    },
    {
      "epoch": 0.7082550533058694,
      "grad_norm": 0.6390408873558044,
      "learning_rate": 3.338711133886169e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 25170816,
      "step": 15645
    },
    {
      "epoch": 0.7084814051925122,
      "grad_norm": 1.0491416454315186,
      "learning_rate": 3.3377862179948064e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 25178816,
      "step": 15650
    },
    {
      "epoch": 0.7087077570791552,
      "grad_norm": 0.6861845850944519,
      "learning_rate": 3.336861172906281e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 25186464,
      "step": 15655
    },
    {
      "epoch": 0.7089341089657982,
      "grad_norm": 0.39936015009880066,
      "learning_rate": 3.335935998763245e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 25194976,
      "step": 15660
    },
    {
      "epoch": 0.7091604608524412,
      "grad_norm": 0.2972133457660675,
      "learning_rate": 3.3350106957083744e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 25203136,
      "step": 15665
    },
    {
      "epoch": 0.7093868127390842,
      "grad_norm": 0.6166517734527588,
      "learning_rate": 3.33408526388436e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 25212000,
      "step": 15670
    },
    {
      "epoch": 0.7096131646257272,
      "grad_norm": 1.1221320629119873,
      "learning_rate": 3.3331597034339166e-05,
      "loss": 0.1653,
      "num_input_tokens_seen": 25219968,
      "step": 15675
    },
    {
      "epoch": 0.7098395165123701,
      "grad_norm": 0.60939621925354,
      "learning_rate": 3.3322340144997764e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 25228192,
      "step": 15680
    },
    {
      "epoch": 0.7100658683990131,
      "grad_norm": 0.919014036655426,
      "learning_rate": 3.331308197224693e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 25236832,
      "step": 15685
    },
    {
      "epoch": 0.7102922202856561,
      "grad_norm": 0.559775710105896,
      "learning_rate": 3.330382251751438e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 25244416,
      "step": 15690
    },
    {
      "epoch": 0.7105185721722991,
      "grad_norm": 0.8180041909217834,
      "learning_rate": 3.3294561782228054e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 25252096,
      "step": 15695
    },
    {
      "epoch": 0.7107449240589421,
      "grad_norm": 0.7638928294181824,
      "learning_rate": 3.328529976781607e-05,
      "loss": 0.2077,
      "num_input_tokens_seen": 25260512,
      "step": 15700
    },
    {
      "epoch": 0.710971275945585,
      "grad_norm": 1.1097058057785034,
      "learning_rate": 3.327603647570673e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 25268896,
      "step": 15705
    },
    {
      "epoch": 0.711197627832228,
      "grad_norm": 0.8637230396270752,
      "learning_rate": 3.326677190732857e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 25277216,
      "step": 15710
    },
    {
      "epoch": 0.711423979718871,
      "grad_norm": 0.4207562208175659,
      "learning_rate": 3.325750606411029e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 25285184,
      "step": 15715
    },
    {
      "epoch": 0.7116503316055139,
      "grad_norm": 0.9980641603469849,
      "learning_rate": 3.3248238947480804e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 25292480,
      "step": 15720
    },
    {
      "epoch": 0.7118766834921569,
      "grad_norm": 0.4773097336292267,
      "learning_rate": 3.323897055886922e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 25299840,
      "step": 15725
    },
    {
      "epoch": 0.7121030353787999,
      "grad_norm": 0.6148668527603149,
      "learning_rate": 3.322970089970484e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 25307936,
      "step": 15730
    },
    {
      "epoch": 0.7123293872654428,
      "grad_norm": 0.6055189967155457,
      "learning_rate": 3.3220429971417165e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 25315872,
      "step": 15735
    },
    {
      "epoch": 0.7125557391520858,
      "grad_norm": 0.5122420191764832,
      "learning_rate": 3.321115777543588e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 25324928,
      "step": 15740
    },
    {
      "epoch": 0.7127820910387288,
      "grad_norm": 0.9703056812286377,
      "learning_rate": 3.320188431319088e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 25332768,
      "step": 15745
    },
    {
      "epoch": 0.7130084429253718,
      "grad_norm": 0.4132006764411926,
      "learning_rate": 3.319260958611224e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 25340608,
      "step": 15750
    },
    {
      "epoch": 0.7132347948120148,
      "grad_norm": 0.8526522517204285,
      "learning_rate": 3.3183333595630256e-05,
      "loss": 0.1316,
      "num_input_tokens_seen": 25348128,
      "step": 15755
    },
    {
      "epoch": 0.7134611466986578,
      "grad_norm": 0.6014826893806458,
      "learning_rate": 3.317405634317538e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 25356192,
      "step": 15760
    },
    {
      "epoch": 0.7136874985853007,
      "grad_norm": 0.6089156866073608,
      "learning_rate": 3.3164777830178315e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 25364416,
      "step": 15765
    },
    {
      "epoch": 0.7139138504719437,
      "grad_norm": 0.6701650619506836,
      "learning_rate": 3.315549805806989e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 25372864,
      "step": 15770
    },
    {
      "epoch": 0.7141402023585867,
      "grad_norm": 0.4416217505931854,
      "learning_rate": 3.314621702828118e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 25380768,
      "step": 15775
    },
    {
      "epoch": 0.7143665542452297,
      "grad_norm": 0.636182427406311,
      "learning_rate": 3.313693474224342e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 25388864,
      "step": 15780
    },
    {
      "epoch": 0.7145929061318727,
      "grad_norm": 0.9425904750823975,
      "learning_rate": 3.312765120138809e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 25396320,
      "step": 15785
    },
    {
      "epoch": 0.7148192580185155,
      "grad_norm": 0.5903092622756958,
      "learning_rate": 3.311836640714679e-05,
      "loss": 0.1641,
      "num_input_tokens_seen": 25404160,
      "step": 15790
    },
    {
      "epoch": 0.7150456099051585,
      "grad_norm": 1.0187904834747314,
      "learning_rate": 3.310908036095137e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 25412672,
      "step": 15795
    },
    {
      "epoch": 0.7152719617918015,
      "grad_norm": 0.6060123443603516,
      "learning_rate": 3.309979306423386e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 25420512,
      "step": 15800
    },
    {
      "epoch": 0.7152719617918015,
      "eval_loss": 0.1509702205657959,
      "eval_runtime": 404.8152,
      "eval_samples_per_second": 97.01,
      "eval_steps_per_second": 24.253,
      "num_input_tokens_seen": 25420512,
      "step": 15800
    },
    {
      "epoch": 0.7154983136784445,
      "grad_norm": 0.8484069108963013,
      "learning_rate": 3.309050451842647e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 25429312,
      "step": 15805
    },
    {
      "epoch": 0.7157246655650875,
      "grad_norm": 0.833393394947052,
      "learning_rate": 3.3081214724961604e-05,
      "loss": 0.1691,
      "num_input_tokens_seen": 25437024,
      "step": 15810
    },
    {
      "epoch": 0.7159510174517305,
      "grad_norm": 0.6639107465744019,
      "learning_rate": 3.307192368527188e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 25445312,
      "step": 15815
    },
    {
      "epoch": 0.7161773693383734,
      "grad_norm": 0.5711321830749512,
      "learning_rate": 3.306263140079008e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 25453632,
      "step": 15820
    },
    {
      "epoch": 0.7164037212250164,
      "grad_norm": 0.7115955948829651,
      "learning_rate": 3.30533378729492e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 25461632,
      "step": 15825
    },
    {
      "epoch": 0.7166300731116594,
      "grad_norm": 0.5549256801605225,
      "learning_rate": 3.304404310318242e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 25469504,
      "step": 15830
    },
    {
      "epoch": 0.7168564249983024,
      "grad_norm": 0.5178081393241882,
      "learning_rate": 3.3034747092923105e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 25477376,
      "step": 15835
    },
    {
      "epoch": 0.7170827768849454,
      "grad_norm": 0.49475419521331787,
      "learning_rate": 3.3025449843604806e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 25485408,
      "step": 15840
    },
    {
      "epoch": 0.7173091287715883,
      "grad_norm": 0.43340298533439636,
      "learning_rate": 3.30161513566613e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 25493440,
      "step": 15845
    },
    {
      "epoch": 0.7175354806582313,
      "grad_norm": 0.3037101626396179,
      "learning_rate": 3.3006851633526506e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 25501312,
      "step": 15850
    },
    {
      "epoch": 0.7177618325448742,
      "grad_norm": 0.46513575315475464,
      "learning_rate": 3.2997550675634584e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 25509760,
      "step": 15855
    },
    {
      "epoch": 0.7179881844315172,
      "grad_norm": 0.320393443107605,
      "learning_rate": 3.2988248484419825e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 25516960,
      "step": 15860
    },
    {
      "epoch": 0.7182145363181602,
      "grad_norm": 0.38967275619506836,
      "learning_rate": 3.2978945061316776e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 25524896,
      "step": 15865
    },
    {
      "epoch": 0.7184408882048032,
      "grad_norm": 0.3412967324256897,
      "learning_rate": 3.296964040776013e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 25533216,
      "step": 15870
    },
    {
      "epoch": 0.7186672400914461,
      "grad_norm": 1.717072606086731,
      "learning_rate": 3.296033452518478e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 25541152,
      "step": 15875
    },
    {
      "epoch": 0.7188935919780891,
      "grad_norm": 0.7821164131164551,
      "learning_rate": 3.2951027415025806e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 25549056,
      "step": 15880
    },
    {
      "epoch": 0.7191199438647321,
      "grad_norm": 1.1109473705291748,
      "learning_rate": 3.294171907871849e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 25557216,
      "step": 15885
    },
    {
      "epoch": 0.7193462957513751,
      "grad_norm": 0.4032238721847534,
      "learning_rate": 3.293240951769828e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 25565280,
      "step": 15890
    },
    {
      "epoch": 0.7195726476380181,
      "grad_norm": 0.4267508089542389,
      "learning_rate": 3.2923098733400846e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 25573696,
      "step": 15895
    },
    {
      "epoch": 0.7197989995246611,
      "grad_norm": 0.7258181571960449,
      "learning_rate": 3.291378672726202e-05,
      "loss": 0.1877,
      "num_input_tokens_seen": 25581344,
      "step": 15900
    },
    {
      "epoch": 0.720025351411304,
      "grad_norm": 0.6156032085418701,
      "learning_rate": 3.2904473500717824e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 25589536,
      "step": 15905
    },
    {
      "epoch": 0.720251703297947,
      "grad_norm": 0.5459427237510681,
      "learning_rate": 3.289515905520449e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 25598016,
      "step": 15910
    },
    {
      "epoch": 0.72047805518459,
      "grad_norm": 0.6336131691932678,
      "learning_rate": 3.288584339215841e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 25606112,
      "step": 15915
    },
    {
      "epoch": 0.720704407071233,
      "grad_norm": 0.6852145195007324,
      "learning_rate": 3.287652651301617e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 25614624,
      "step": 15920
    },
    {
      "epoch": 0.720930758957876,
      "grad_norm": 0.43297305703163147,
      "learning_rate": 3.286720841921457e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 25622560,
      "step": 15925
    },
    {
      "epoch": 0.7211571108445188,
      "grad_norm": 0.7688985466957092,
      "learning_rate": 3.285788911219056e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 25631136,
      "step": 15930
    },
    {
      "epoch": 0.7213834627311618,
      "grad_norm": 0.5584052801132202,
      "learning_rate": 3.284856859338131e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 25638944,
      "step": 15935
    },
    {
      "epoch": 0.7216098146178048,
      "grad_norm": 0.6311877369880676,
      "learning_rate": 3.283924686422414e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 25647776,
      "step": 15940
    },
    {
      "epoch": 0.7218361665044478,
      "grad_norm": 0.5611324906349182,
      "learning_rate": 3.282992392615659e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 25656224,
      "step": 15945
    },
    {
      "epoch": 0.7220625183910908,
      "grad_norm": 0.5167104601860046,
      "learning_rate": 3.282059978061638e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 25663904,
      "step": 15950
    },
    {
      "epoch": 0.7222888702777338,
      "grad_norm": 0.3068002462387085,
      "learning_rate": 3.28112744290414e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 25672128,
      "step": 15955
    },
    {
      "epoch": 0.7225152221643767,
      "grad_norm": 0.7767694592475891,
      "learning_rate": 3.280194787286974e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 25680928,
      "step": 15960
    },
    {
      "epoch": 0.7227415740510197,
      "grad_norm": 0.8122792840003967,
      "learning_rate": 3.2792620113539674e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 25689216,
      "step": 15965
    },
    {
      "epoch": 0.7229679259376627,
      "grad_norm": 0.7328422665596008,
      "learning_rate": 3.278329115248966e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 25697312,
      "step": 15970
    },
    {
      "epoch": 0.7231942778243057,
      "grad_norm": 0.6218723058700562,
      "learning_rate": 3.277396099115834e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 25705792,
      "step": 15975
    },
    {
      "epoch": 0.7234206297109487,
      "grad_norm": 0.46003711223602295,
      "learning_rate": 3.276462963098454e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 25714080,
      "step": 15980
    },
    {
      "epoch": 0.7236469815975917,
      "grad_norm": 0.5352490544319153,
      "learning_rate": 3.275529707340728e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 25722880,
      "step": 15985
    },
    {
      "epoch": 0.7238733334842345,
      "grad_norm": 0.7695570588111877,
      "learning_rate": 3.274596331986574e-05,
      "loss": 0.182,
      "num_input_tokens_seen": 25730976,
      "step": 15990
    },
    {
      "epoch": 0.7240996853708775,
      "grad_norm": 0.553909420967102,
      "learning_rate": 3.273662837179932e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 25738720,
      "step": 15995
    },
    {
      "epoch": 0.7243260372575205,
      "grad_norm": 0.9517850279808044,
      "learning_rate": 3.272729223064758e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 25746912,
      "step": 16000
    },
    {
      "epoch": 0.7243260372575205,
      "eval_loss": 0.15068061649799347,
      "eval_runtime": 404.9374,
      "eval_samples_per_second": 96.98,
      "eval_steps_per_second": 24.246,
      "num_input_tokens_seen": 25746912,
      "step": 16000
    },
    {
      "epoch": 0.7245523891441635,
      "grad_norm": 0.9182016253471375,
      "learning_rate": 3.2717954897850264e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 25755008,
      "step": 16005
    },
    {
      "epoch": 0.7247787410308065,
      "grad_norm": 0.6410081386566162,
      "learning_rate": 3.270861637484733e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 25762752,
      "step": 16010
    },
    {
      "epoch": 0.7250050929174494,
      "grad_norm": 0.4600309431552887,
      "learning_rate": 3.2699276663078867e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 25770720,
      "step": 16015
    },
    {
      "epoch": 0.7252314448040924,
      "grad_norm": 0.43933236598968506,
      "learning_rate": 3.268993576398519e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 25778496,
      "step": 16020
    },
    {
      "epoch": 0.7254577966907354,
      "grad_norm": 0.5936673283576965,
      "learning_rate": 3.268059367900678e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 25785856,
      "step": 16025
    },
    {
      "epoch": 0.7256841485773784,
      "grad_norm": 0.8210378289222717,
      "learning_rate": 3.26712504095843e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 25794720,
      "step": 16030
    },
    {
      "epoch": 0.7259105004640214,
      "grad_norm": 0.3904089331626892,
      "learning_rate": 3.2661905957158615e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 25802496,
      "step": 16035
    },
    {
      "epoch": 0.7261368523506644,
      "grad_norm": 0.28991732001304626,
      "learning_rate": 3.2652560323170734e-05,
      "loss": 0.1776,
      "num_input_tokens_seen": 25810784,
      "step": 16040
    },
    {
      "epoch": 0.7263632042373073,
      "grad_norm": 0.5228879451751709,
      "learning_rate": 3.264321350906189e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 25819168,
      "step": 16045
    },
    {
      "epoch": 0.7265895561239503,
      "grad_norm": 0.6603827476501465,
      "learning_rate": 3.263386551627346e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 25826880,
      "step": 16050
    },
    {
      "epoch": 0.7268159080105933,
      "grad_norm": 0.9570427536964417,
      "learning_rate": 3.2624516346247055e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 25835456,
      "step": 16055
    },
    {
      "epoch": 0.7270422598972363,
      "grad_norm": 0.5731454491615295,
      "learning_rate": 3.2615166000424404e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 25843744,
      "step": 16060
    },
    {
      "epoch": 0.7272686117838792,
      "grad_norm": 0.7399309277534485,
      "learning_rate": 3.260581448024745e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 25851840,
      "step": 16065
    },
    {
      "epoch": 0.7274949636705222,
      "grad_norm": 0.9541402459144592,
      "learning_rate": 3.2596461787158335e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 25859936,
      "step": 16070
    },
    {
      "epoch": 0.7277213155571651,
      "grad_norm": 0.6807599067687988,
      "learning_rate": 3.258710792259934e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 25867584,
      "step": 16075
    },
    {
      "epoch": 0.7279476674438081,
      "grad_norm": 0.4395805299282074,
      "learning_rate": 3.257775288801296e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 25875808,
      "step": 16080
    },
    {
      "epoch": 0.7281740193304511,
      "grad_norm": 0.4921571612358093,
      "learning_rate": 3.256839668484186e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 25883744,
      "step": 16085
    },
    {
      "epoch": 0.7284003712170941,
      "grad_norm": 0.4135526418685913,
      "learning_rate": 3.255903931452888e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 25891840,
      "step": 16090
    },
    {
      "epoch": 0.7286267231037371,
      "grad_norm": 0.5286211967468262,
      "learning_rate": 3.2549680778517045e-05,
      "loss": 0.1705,
      "num_input_tokens_seen": 25899872,
      "step": 16095
    },
    {
      "epoch": 0.72885307499038,
      "grad_norm": 0.5964949131011963,
      "learning_rate": 3.2540321078249556e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 25907936,
      "step": 16100
    },
    {
      "epoch": 0.729079426877023,
      "grad_norm": 0.6563379168510437,
      "learning_rate": 3.2530960215169795e-05,
      "loss": 0.1681,
      "num_input_tokens_seen": 25916224,
      "step": 16105
    },
    {
      "epoch": 0.729305778763666,
      "grad_norm": 0.6040756702423096,
      "learning_rate": 3.2521598190721345e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 25924192,
      "step": 16110
    },
    {
      "epoch": 0.729532130650309,
      "grad_norm": 1.2981520891189575,
      "learning_rate": 3.251223500634792e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 25931904,
      "step": 16115
    },
    {
      "epoch": 0.729758482536952,
      "grad_norm": 0.6788409948348999,
      "learning_rate": 3.2502870663493445e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 25939552,
      "step": 16120
    },
    {
      "epoch": 0.729984834423595,
      "grad_norm": 0.7788093686103821,
      "learning_rate": 3.249350516360203e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 25947712,
      "step": 16125
    },
    {
      "epoch": 0.7302111863102378,
      "grad_norm": 0.7405436038970947,
      "learning_rate": 3.248413850811797e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 25955648,
      "step": 16130
    },
    {
      "epoch": 0.7304375381968808,
      "grad_norm": 0.3250063359737396,
      "learning_rate": 3.2474770698485677e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 25964160,
      "step": 16135
    },
    {
      "epoch": 0.7306638900835238,
      "grad_norm": 0.7035131454467773,
      "learning_rate": 3.246540173614983e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 25972384,
      "step": 16140
    },
    {
      "epoch": 0.7308902419701668,
      "grad_norm": 0.7898593544960022,
      "learning_rate": 3.2456031622555197e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 25980224,
      "step": 16145
    },
    {
      "epoch": 0.7311165938568098,
      "grad_norm": 0.34793323278427124,
      "learning_rate": 3.2446660359146794e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 25988320,
      "step": 16150
    },
    {
      "epoch": 0.7313429457434528,
      "grad_norm": 0.4884036183357239,
      "learning_rate": 3.2437287947369786e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 25996544,
      "step": 16155
    },
    {
      "epoch": 0.7315692976300957,
      "grad_norm": 0.4006035327911377,
      "learning_rate": 3.2427914388669525e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 26004256,
      "step": 16160
    },
    {
      "epoch": 0.7317956495167387,
      "grad_norm": 0.8243975639343262,
      "learning_rate": 3.241853968449151e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 26012832,
      "step": 16165
    },
    {
      "epoch": 0.7320220014033817,
      "grad_norm": 0.5367249846458435,
      "learning_rate": 3.240916383628144e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 26021440,
      "step": 16170
    },
    {
      "epoch": 0.7322483532900247,
      "grad_norm": 0.3357580006122589,
      "learning_rate": 3.239978684548521e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 26029472,
      "step": 16175
    },
    {
      "epoch": 0.7324747051766677,
      "grad_norm": 1.1496679782867432,
      "learning_rate": 3.239040871354885e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 26037600,
      "step": 16180
    },
    {
      "epoch": 0.7327010570633106,
      "grad_norm": 0.4824647605419159,
      "learning_rate": 3.2381029441918596e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 26045152,
      "step": 16185
    },
    {
      "epoch": 0.7329274089499536,
      "grad_norm": 0.7558034062385559,
      "learning_rate": 3.2371649032040845e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 26053376,
      "step": 16190
    },
    {
      "epoch": 0.7331537608365966,
      "grad_norm": 1.0665254592895508,
      "learning_rate": 3.2362267485362174e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 26060960,
      "step": 16195
    },
    {
      "epoch": 0.7333801127232396,
      "grad_norm": 0.8273358345031738,
      "learning_rate": 3.235288480332934e-05,
      "loss": 0.1514,
      "num_input_tokens_seen": 26070048,
      "step": 16200
    },
    {
      "epoch": 0.7333801127232396,
      "eval_loss": 0.15163125097751617,
      "eval_runtime": 405.5198,
      "eval_samples_per_second": 96.841,
      "eval_steps_per_second": 24.211,
      "num_input_tokens_seen": 26070048,
      "step": 16200
    },
    {
      "epoch": 0.7336064646098825,
      "grad_norm": 1.1677006483078003,
      "learning_rate": 3.234350098738927e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 26077952,
      "step": 16205
    },
    {
      "epoch": 0.7338328164965255,
      "grad_norm": 1.002488374710083,
      "learning_rate": 3.233411603898906e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 26085728,
      "step": 16210
    },
    {
      "epoch": 0.7340591683831684,
      "grad_norm": 0.4773003160953522,
      "learning_rate": 3.232472995957599e-05,
      "loss": 0.1258,
      "num_input_tokens_seen": 26093696,
      "step": 16215
    },
    {
      "epoch": 0.7342855202698114,
      "grad_norm": 0.5375311970710754,
      "learning_rate": 3.231534275059751e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 26101248,
      "step": 16220
    },
    {
      "epoch": 0.7345118721564544,
      "grad_norm": 1.106979250907898,
      "learning_rate": 3.230595441350125e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 26109024,
      "step": 16225
    },
    {
      "epoch": 0.7347382240430974,
      "grad_norm": 1.6625726222991943,
      "learning_rate": 3.2296564949735e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 26116448,
      "step": 16230
    },
    {
      "epoch": 0.7349645759297404,
      "grad_norm": 0.7997708320617676,
      "learning_rate": 3.228717436074675e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 26124512,
      "step": 16235
    },
    {
      "epoch": 0.7351909278163834,
      "grad_norm": 0.9070101380348206,
      "learning_rate": 3.227778264798463e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 26132640,
      "step": 16240
    },
    {
      "epoch": 0.7354172797030263,
      "grad_norm": 0.8132191896438599,
      "learning_rate": 3.226838981289698e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 26140960,
      "step": 16245
    },
    {
      "epoch": 0.7356436315896693,
      "grad_norm": 0.8028194308280945,
      "learning_rate": 3.225899585693227e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 26148864,
      "step": 16250
    },
    {
      "epoch": 0.7358699834763123,
      "grad_norm": 0.5128571391105652,
      "learning_rate": 3.224960078153918e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 26157664,
      "step": 16255
    },
    {
      "epoch": 0.7360963353629553,
      "grad_norm": 1.5343812704086304,
      "learning_rate": 3.224020458816655e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 26165696,
      "step": 16260
    },
    {
      "epoch": 0.7363226872495983,
      "grad_norm": 0.5125400424003601,
      "learning_rate": 3.223080727826337e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 26173728,
      "step": 16265
    },
    {
      "epoch": 0.7365490391362411,
      "grad_norm": 1.1036278009414673,
      "learning_rate": 3.222140885327885e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 26181984,
      "step": 16270
    },
    {
      "epoch": 0.7367753910228841,
      "grad_norm": 0.34613096714019775,
      "learning_rate": 3.221200931466234e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 26190048,
      "step": 16275
    },
    {
      "epoch": 0.7370017429095271,
      "grad_norm": 0.6562451720237732,
      "learning_rate": 3.220260866386336e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 26197792,
      "step": 16280
    },
    {
      "epoch": 0.7372280947961701,
      "grad_norm": 0.5569134950637817,
      "learning_rate": 3.21932069023316e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 26205792,
      "step": 16285
    },
    {
      "epoch": 0.7374544466828131,
      "grad_norm": 0.871982216835022,
      "learning_rate": 3.218380403151695e-05,
      "loss": 0.0969,
      "num_input_tokens_seen": 26214176,
      "step": 16290
    },
    {
      "epoch": 0.7376807985694561,
      "grad_norm": 0.47488030791282654,
      "learning_rate": 3.217440005286943e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 26221760,
      "step": 16295
    },
    {
      "epoch": 0.737907150456099,
      "grad_norm": 0.677266001701355,
      "learning_rate": 3.216499496783928e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 26229824,
      "step": 16300
    },
    {
      "epoch": 0.738133502342742,
      "grad_norm": 0.5938441753387451,
      "learning_rate": 3.2155588777876856e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 26237952,
      "step": 16305
    },
    {
      "epoch": 0.738359854229385,
      "grad_norm": 0.5020385384559631,
      "learning_rate": 3.214618148443273e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 26245824,
      "step": 16310
    },
    {
      "epoch": 0.738586206116028,
      "grad_norm": 0.6934816241264343,
      "learning_rate": 3.2136773088957595e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 26253408,
      "step": 16315
    },
    {
      "epoch": 0.738812558002671,
      "grad_norm": 0.7645438313484192,
      "learning_rate": 3.2127363592902374e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 26262016,
      "step": 16320
    },
    {
      "epoch": 0.739038909889314,
      "grad_norm": 1.2627071142196655,
      "learning_rate": 3.211795299771812e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 26269696,
      "step": 16325
    },
    {
      "epoch": 0.7392652617759569,
      "grad_norm": 0.4359349012374878,
      "learning_rate": 3.210854130485605e-05,
      "loss": 0.1413,
      "num_input_tokens_seen": 26278528,
      "step": 16330
    },
    {
      "epoch": 0.7394916136625999,
      "grad_norm": 0.673775851726532,
      "learning_rate": 3.209912851576759e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 26286208,
      "step": 16335
    },
    {
      "epoch": 0.7397179655492429,
      "grad_norm": 0.39114639163017273,
      "learning_rate": 3.208971463190431e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 26294112,
      "step": 16340
    },
    {
      "epoch": 0.7399443174358858,
      "grad_norm": 0.6553220152854919,
      "learning_rate": 3.208029965471793e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 26301984,
      "step": 16345
    },
    {
      "epoch": 0.7401706693225288,
      "grad_norm": 0.7402985095977783,
      "learning_rate": 3.2070883585660364e-05,
      "loss": 0.1598,
      "num_input_tokens_seen": 26310144,
      "step": 16350
    },
    {
      "epoch": 0.7403970212091717,
      "grad_norm": 0.6799684166908264,
      "learning_rate": 3.20614664261837e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 26318240,
      "step": 16355
    },
    {
      "epoch": 0.7406233730958147,
      "grad_norm": 0.8112844228744507,
      "learning_rate": 3.205204817774016e-05,
      "loss": 0.1143,
      "num_input_tokens_seen": 26326432,
      "step": 16360
    },
    {
      "epoch": 0.7408497249824577,
      "grad_norm": 0.8144729733467102,
      "learning_rate": 3.204262884178218e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 26334144,
      "step": 16365
    },
    {
      "epoch": 0.7410760768691007,
      "grad_norm": 0.6466651558876038,
      "learning_rate": 3.2033208419762314e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 26341664,
      "step": 16370
    },
    {
      "epoch": 0.7413024287557437,
      "grad_norm": 1.411866545677185,
      "learning_rate": 3.2023786913133344e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 26349312,
      "step": 16375
    },
    {
      "epoch": 0.7415287806423867,
      "grad_norm": 0.6355913877487183,
      "learning_rate": 3.201436432334816e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 26357184,
      "step": 16380
    },
    {
      "epoch": 0.7417551325290296,
      "grad_norm": 1.220456838607788,
      "learning_rate": 3.2004940651859844e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 26365440,
      "step": 16385
    },
    {
      "epoch": 0.7419814844156726,
      "grad_norm": 0.8891407251358032,
      "learning_rate": 3.1995515900121655e-05,
      "loss": 0.2184,
      "num_input_tokens_seen": 26373184,
      "step": 16390
    },
    {
      "epoch": 0.7422078363023156,
      "grad_norm": 0.44358348846435547,
      "learning_rate": 3.1986090069587e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 26381248,
      "step": 16395
    },
    {
      "epoch": 0.7424341881889586,
      "grad_norm": 0.5736426711082458,
      "learning_rate": 3.1976663161709466e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 26389952,
      "step": 16400
    },
    {
      "epoch": 0.7424341881889586,
      "eval_loss": 0.15023045241832733,
      "eval_runtime": 404.9073,
      "eval_samples_per_second": 96.988,
      "eval_steps_per_second": 24.248,
      "num_input_tokens_seen": 26389952,
      "step": 16400
    },
    {
      "epoch": 0.7426605400756016,
      "grad_norm": 0.748035192489624,
      "learning_rate": 3.196723517794279e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 26397696,
      "step": 16405
    },
    {
      "epoch": 0.7428868919622446,
      "grad_norm": 0.9180954694747925,
      "learning_rate": 3.19578061197409e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 26405472,
      "step": 16410
    },
    {
      "epoch": 0.7431132438488874,
      "grad_norm": 1.082252025604248,
      "learning_rate": 3.194837598855787e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 26413088,
      "step": 16415
    },
    {
      "epoch": 0.7433395957355304,
      "grad_norm": 0.8463055491447449,
      "learning_rate": 3.193894478584794e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 26421440,
      "step": 16420
    },
    {
      "epoch": 0.7435659476221734,
      "grad_norm": 0.37035393714904785,
      "learning_rate": 3.192951251306553e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 26429408,
      "step": 16425
    },
    {
      "epoch": 0.7437922995088164,
      "grad_norm": 0.6365521550178528,
      "learning_rate": 3.192007917166521e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 26437312,
      "step": 16430
    },
    {
      "epoch": 0.7440186513954594,
      "grad_norm": 1.0417853593826294,
      "learning_rate": 3.191064476310171e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 26445312,
      "step": 16435
    },
    {
      "epoch": 0.7442450032821023,
      "grad_norm": 0.5546200275421143,
      "learning_rate": 3.1901209288829944e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 26453216,
      "step": 16440
    },
    {
      "epoch": 0.7444713551687453,
      "grad_norm": 0.9149183630943298,
      "learning_rate": 3.1891772750304985e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 26461600,
      "step": 16445
    },
    {
      "epoch": 0.7446977070553883,
      "grad_norm": 1.3695510625839233,
      "learning_rate": 3.188233514898206e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 26469216,
      "step": 16450
    },
    {
      "epoch": 0.7449240589420313,
      "grad_norm": 0.38912931084632874,
      "learning_rate": 3.187289648631657e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 26476832,
      "step": 16455
    },
    {
      "epoch": 0.7451504108286743,
      "grad_norm": 0.3079336881637573,
      "learning_rate": 3.186345676376406e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 26484672,
      "step": 16460
    },
    {
      "epoch": 0.7453767627153173,
      "grad_norm": 0.47752702236175537,
      "learning_rate": 3.1854015982780275e-05,
      "loss": 0.0973,
      "num_input_tokens_seen": 26492320,
      "step": 16465
    },
    {
      "epoch": 0.7456031146019602,
      "grad_norm": 0.38316038250923157,
      "learning_rate": 3.1844574144821084e-05,
      "loss": 0.1856,
      "num_input_tokens_seen": 26500864,
      "step": 16470
    },
    {
      "epoch": 0.7458294664886032,
      "grad_norm": 0.9269040822982788,
      "learning_rate": 3.1835131251342554e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 26508672,
      "step": 16475
    },
    {
      "epoch": 0.7460558183752461,
      "grad_norm": 0.3240070641040802,
      "learning_rate": 3.182568730380089e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 26517120,
      "step": 16480
    },
    {
      "epoch": 0.7462821702618891,
      "grad_norm": 0.50316321849823,
      "learning_rate": 3.181624230365245e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 26525120,
      "step": 16485
    },
    {
      "epoch": 0.7465085221485321,
      "grad_norm": 1.000488519668579,
      "learning_rate": 3.180679625235381e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 26533408,
      "step": 16490
    },
    {
      "epoch": 0.7467348740351751,
      "grad_norm": 0.4339582026004791,
      "learning_rate": 3.1797349151361646e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 26541664,
      "step": 16495
    },
    {
      "epoch": 0.746961225921818,
      "grad_norm": 0.2572067975997925,
      "learning_rate": 3.178790100213281e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 26550304,
      "step": 16500
    },
    {
      "epoch": 0.747187577808461,
      "grad_norm": 0.5211760997772217,
      "learning_rate": 3.1778451806124346e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 26558176,
      "step": 16505
    },
    {
      "epoch": 0.747413929695104,
      "grad_norm": 0.8294855356216431,
      "learning_rate": 3.176900156479342e-05,
      "loss": 0.1093,
      "num_input_tokens_seen": 26566208,
      "step": 16510
    },
    {
      "epoch": 0.747640281581747,
      "grad_norm": 0.44295334815979004,
      "learning_rate": 3.17595502795974e-05,
      "loss": 0.1281,
      "num_input_tokens_seen": 26574112,
      "step": 16515
    },
    {
      "epoch": 0.74786663346839,
      "grad_norm": 0.5614180564880371,
      "learning_rate": 3.175009795199377e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 26581952,
      "step": 16520
    },
    {
      "epoch": 0.7480929853550329,
      "grad_norm": 0.4833396077156067,
      "learning_rate": 3.1740644583440224e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 26590368,
      "step": 16525
    },
    {
      "epoch": 0.7483193372416759,
      "grad_norm": 0.45746591687202454,
      "learning_rate": 3.173119017539457e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 26598496,
      "step": 16530
    },
    {
      "epoch": 0.7485456891283189,
      "grad_norm": 0.5053521990776062,
      "learning_rate": 3.172173472931479e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 26606368,
      "step": 16535
    },
    {
      "epoch": 0.7487720410149619,
      "grad_norm": 0.7590806484222412,
      "learning_rate": 3.1712278246659055e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 26614752,
      "step": 16540
    },
    {
      "epoch": 0.7489983929016049,
      "grad_norm": 0.5509451031684875,
      "learning_rate": 3.170282072888566e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 26622400,
      "step": 16545
    },
    {
      "epoch": 0.7492247447882479,
      "grad_norm": 0.4758554995059967,
      "learning_rate": 3.169336217745307e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 26630304,
      "step": 16550
    },
    {
      "epoch": 0.7494510966748907,
      "grad_norm": 1.137190580368042,
      "learning_rate": 3.1683902593819924e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 26637984,
      "step": 16555
    },
    {
      "epoch": 0.7496774485615337,
      "grad_norm": 1.016136646270752,
      "learning_rate": 3.1674441979445e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 26646016,
      "step": 16560
    },
    {
      "epoch": 0.7499038004481767,
      "grad_norm": 0.428372859954834,
      "learning_rate": 3.166498033578725e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 26653632,
      "step": 16565
    },
    {
      "epoch": 0.7501301523348197,
      "grad_norm": 1.0910955667495728,
      "learning_rate": 3.165551766430578e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 26660928,
      "step": 16570
    },
    {
      "epoch": 0.7503565042214627,
      "grad_norm": 0.34917035698890686,
      "learning_rate": 3.164605396645984e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 26668128,
      "step": 16575
    },
    {
      "epoch": 0.7505828561081057,
      "grad_norm": 1.0853341817855835,
      "learning_rate": 3.163658924370886e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 26675712,
      "step": 16580
    },
    {
      "epoch": 0.7508092079947486,
      "grad_norm": 0.7465262413024902,
      "learning_rate": 3.1627123497512415e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 26683136,
      "step": 16585
    },
    {
      "epoch": 0.7510355598813916,
      "grad_norm": 0.8435594439506531,
      "learning_rate": 3.1617656729330245e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 26690528,
      "step": 16590
    },
    {
      "epoch": 0.7512619117680346,
      "grad_norm": 0.5973437428474426,
      "learning_rate": 3.1608188940622255e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 26698144,
      "step": 16595
    },
    {
      "epoch": 0.7514882636546776,
      "grad_norm": 0.41259387135505676,
      "learning_rate": 3.159872013284847e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 26706080,
      "step": 16600
    },
    {
      "epoch": 0.7514882636546776,
      "eval_loss": 0.1499524712562561,
      "eval_runtime": 404.3962,
      "eval_samples_per_second": 97.11,
      "eval_steps_per_second": 24.278,
      "num_input_tokens_seen": 26706080,
      "step": 16600
    },
    {
      "epoch": 0.7517146155413206,
      "grad_norm": 1.7818080186843872,
      "learning_rate": 3.1589250307469134e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 26714240,
      "step": 16605
    },
    {
      "epoch": 0.7519409674279635,
      "grad_norm": 0.8368114829063416,
      "learning_rate": 3.1579779465944586e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 26722560,
      "step": 16610
    },
    {
      "epoch": 0.7521673193146065,
      "grad_norm": 0.7952853441238403,
      "learning_rate": 3.1570307609735363e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 26730912,
      "step": 16615
    },
    {
      "epoch": 0.7523936712012494,
      "grad_norm": 0.7670140266418457,
      "learning_rate": 3.156083474030213e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 26738624,
      "step": 16620
    },
    {
      "epoch": 0.7526200230878924,
      "grad_norm": 0.47446224093437195,
      "learning_rate": 3.155136085910573e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 26745952,
      "step": 16625
    },
    {
      "epoch": 0.7528463749745354,
      "grad_norm": 1.047397494316101,
      "learning_rate": 3.154188596760717e-05,
      "loss": 0.1499,
      "num_input_tokens_seen": 26755104,
      "step": 16630
    },
    {
      "epoch": 0.7530727268611784,
      "grad_norm": 0.30561408400535583,
      "learning_rate": 3.153241006726757e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 26762656,
      "step": 16635
    },
    {
      "epoch": 0.7532990787478213,
      "grad_norm": 0.29482322931289673,
      "learning_rate": 3.152293315954825e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 26770112,
      "step": 16640
    },
    {
      "epoch": 0.7535254306344643,
      "grad_norm": 0.3923913836479187,
      "learning_rate": 3.1513455245910666e-05,
      "loss": 0.1069,
      "num_input_tokens_seen": 26778016,
      "step": 16645
    },
    {
      "epoch": 0.7537517825211073,
      "grad_norm": 0.9691307544708252,
      "learning_rate": 3.150397632781643e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 26785984,
      "step": 16650
    },
    {
      "epoch": 0.7539781344077503,
      "grad_norm": 0.4357251822948456,
      "learning_rate": 3.149449640672731e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 26793728,
      "step": 16655
    },
    {
      "epoch": 0.7542044862943933,
      "grad_norm": 0.3678920269012451,
      "learning_rate": 3.148501548410523e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 26801696,
      "step": 16660
    },
    {
      "epoch": 0.7544308381810363,
      "grad_norm": 0.3755541145801544,
      "learning_rate": 3.1475533561412256e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 26809952,
      "step": 16665
    },
    {
      "epoch": 0.7546571900676792,
      "grad_norm": 0.5994240045547485,
      "learning_rate": 3.146605064011065e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 26818528,
      "step": 16670
    },
    {
      "epoch": 0.7548835419543222,
      "grad_norm": 0.5483283996582031,
      "learning_rate": 3.145656672166277e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 26827296,
      "step": 16675
    },
    {
      "epoch": 0.7551098938409652,
      "grad_norm": 0.9432874917984009,
      "learning_rate": 3.144708180753116e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 26835200,
      "step": 16680
    },
    {
      "epoch": 0.7553362457276082,
      "grad_norm": 0.6067450642585754,
      "learning_rate": 3.143759589917851e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 26843040,
      "step": 16685
    },
    {
      "epoch": 0.7555625976142512,
      "grad_norm": 1.068743348121643,
      "learning_rate": 3.142810899806768e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 26851008,
      "step": 16690
    },
    {
      "epoch": 0.755788949500894,
      "grad_norm": 0.4683462083339691,
      "learning_rate": 3.141862110566166e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 26858816,
      "step": 16695
    },
    {
      "epoch": 0.756015301387537,
      "grad_norm": 0.5198163986206055,
      "learning_rate": 3.1409132223423606e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 26867232,
      "step": 16700
    },
    {
      "epoch": 0.75624165327418,
      "grad_norm": 0.7486979365348816,
      "learning_rate": 3.139964235281682e-05,
      "loss": 0.1805,
      "num_input_tokens_seen": 26875744,
      "step": 16705
    },
    {
      "epoch": 0.756468005160823,
      "grad_norm": 0.8713725209236145,
      "learning_rate": 3.139015149530476e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 26883616,
      "step": 16710
    },
    {
      "epoch": 0.756694357047466,
      "grad_norm": 0.5767411589622498,
      "learning_rate": 3.1380659652351034e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 26892128,
      "step": 16715
    },
    {
      "epoch": 0.756920708934109,
      "grad_norm": 1.0252772569656372,
      "learning_rate": 3.137116682541941e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 26899488,
      "step": 16720
    },
    {
      "epoch": 0.7571470608207519,
      "grad_norm": 0.986281156539917,
      "learning_rate": 3.136167301597379e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 26907488,
      "step": 16725
    },
    {
      "epoch": 0.7573734127073949,
      "grad_norm": 0.5237967371940613,
      "learning_rate": 3.1352178225478254e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 26915744,
      "step": 16730
    },
    {
      "epoch": 0.7575997645940379,
      "grad_norm": 0.45078763365745544,
      "learning_rate": 3.1342682455396996e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 26923840,
      "step": 16735
    },
    {
      "epoch": 0.7578261164806809,
      "grad_norm": 0.37360355257987976,
      "learning_rate": 3.133318570719441e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 26932192,
      "step": 16740
    },
    {
      "epoch": 0.7580524683673239,
      "grad_norm": 0.5688714385032654,
      "learning_rate": 3.132368798233499e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 26940448,
      "step": 16745
    },
    {
      "epoch": 0.7582788202539669,
      "grad_norm": 1.3592652082443237,
      "learning_rate": 3.131418928228342e-05,
      "loss": 0.1896,
      "num_input_tokens_seen": 26948352,
      "step": 16750
    },
    {
      "epoch": 0.7585051721406098,
      "grad_norm": 0.6858753561973572,
      "learning_rate": 3.1304689608504514e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 26956864,
      "step": 16755
    },
    {
      "epoch": 0.7587315240272527,
      "grad_norm": 0.4639369249343872,
      "learning_rate": 3.129518896246324e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 26966432,
      "step": 16760
    },
    {
      "epoch": 0.7589578759138957,
      "grad_norm": 0.5705445408821106,
      "learning_rate": 3.128568734562472e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 26974624,
      "step": 16765
    },
    {
      "epoch": 0.7591842278005387,
      "grad_norm": 1.2056429386138916,
      "learning_rate": 3.127618475945421e-05,
      "loss": 0.1855,
      "num_input_tokens_seen": 26982496,
      "step": 16770
    },
    {
      "epoch": 0.7594105796871817,
      "grad_norm": 1.0871473550796509,
      "learning_rate": 3.126668120541715e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 26990368,
      "step": 16775
    },
    {
      "epoch": 0.7596369315738246,
      "grad_norm": 1.0371224880218506,
      "learning_rate": 3.1257176684979096e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 26998400,
      "step": 16780
    },
    {
      "epoch": 0.7598632834604676,
      "grad_norm": 0.5031028389930725,
      "learning_rate": 3.124767119960576e-05,
      "loss": 0.1654,
      "num_input_tokens_seen": 27006336,
      "step": 16785
    },
    {
      "epoch": 0.7600896353471106,
      "grad_norm": 0.9781157374382019,
      "learning_rate": 3.123816475076301e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 27014112,
      "step": 16790
    },
    {
      "epoch": 0.7603159872337536,
      "grad_norm": 0.5039782524108887,
      "learning_rate": 3.122865733991687e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 27021792,
      "step": 16795
    },
    {
      "epoch": 0.7605423391203966,
      "grad_norm": 0.533564031124115,
      "learning_rate": 3.1219148968533486e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 27029344,
      "step": 16800
    },
    {
      "epoch": 0.7605423391203966,
      "eval_loss": 0.150989830493927,
      "eval_runtime": 404.9589,
      "eval_samples_per_second": 96.975,
      "eval_steps_per_second": 24.244,
      "num_input_tokens_seen": 27029344,
      "step": 16800
    },
    {
      "epoch": 0.7607686910070396,
      "grad_norm": 0.8186802268028259,
      "learning_rate": 3.120963963807918e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 27037440,
      "step": 16805
    },
    {
      "epoch": 0.7609950428936825,
      "grad_norm": 0.5554300546646118,
      "learning_rate": 3.12001293500204e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 27045792,
      "step": 16810
    },
    {
      "epoch": 0.7612213947803255,
      "grad_norm": 0.38451626896858215,
      "learning_rate": 3.1190618105823765e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 27053792,
      "step": 16815
    },
    {
      "epoch": 0.7614477466669685,
      "grad_norm": 0.8579367995262146,
      "learning_rate": 3.118110590695603e-05,
      "loss": 0.1145,
      "num_input_tokens_seen": 27061248,
      "step": 16820
    },
    {
      "epoch": 0.7616740985536115,
      "grad_norm": 0.7322536706924438,
      "learning_rate": 3.117159275488407e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 27070304,
      "step": 16825
    },
    {
      "epoch": 0.7619004504402545,
      "grad_norm": 1.1193689107894897,
      "learning_rate": 3.1162078651074956e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 27077856,
      "step": 16830
    },
    {
      "epoch": 0.7621268023268974,
      "grad_norm": 1.597363829612732,
      "learning_rate": 3.1152563596995885e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 27086720,
      "step": 16835
    },
    {
      "epoch": 0.7623531542135403,
      "grad_norm": 0.9247674942016602,
      "learning_rate": 3.1143047594114186e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 27095040,
      "step": 16840
    },
    {
      "epoch": 0.7625795061001833,
      "grad_norm": 0.8466004133224487,
      "learning_rate": 3.113353064389734e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 27102784,
      "step": 16845
    },
    {
      "epoch": 0.7628058579868263,
      "grad_norm": 0.6794431209564209,
      "learning_rate": 3.1124012747812993e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 27110592,
      "step": 16850
    },
    {
      "epoch": 0.7630322098734693,
      "grad_norm": 0.5867159366607666,
      "learning_rate": 3.1114493907328936e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 27118144,
      "step": 16855
    },
    {
      "epoch": 0.7632585617601123,
      "grad_norm": 0.5105831027030945,
      "learning_rate": 3.110497412391306e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 27126080,
      "step": 16860
    },
    {
      "epoch": 0.7634849136467552,
      "grad_norm": 0.7135942578315735,
      "learning_rate": 3.1095453399033466e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 27134048,
      "step": 16865
    },
    {
      "epoch": 0.7637112655333982,
      "grad_norm": 1.6270194053649902,
      "learning_rate": 3.108593173415835e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 27142336,
      "step": 16870
    },
    {
      "epoch": 0.7639376174200412,
      "grad_norm": 0.6157485246658325,
      "learning_rate": 3.107640913075609e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 27150496,
      "step": 16875
    },
    {
      "epoch": 0.7641639693066842,
      "grad_norm": 0.9028968214988708,
      "learning_rate": 3.106688559029517e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 27158816,
      "step": 16880
    },
    {
      "epoch": 0.7643903211933272,
      "grad_norm": 0.9634581208229065,
      "learning_rate": 3.105736111424425e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 27167360,
      "step": 16885
    },
    {
      "epoch": 0.7646166730799702,
      "grad_norm": 0.9245402812957764,
      "learning_rate": 3.1047835704072136e-05,
      "loss": 0.1833,
      "num_input_tokens_seen": 27175008,
      "step": 16890
    },
    {
      "epoch": 0.764843024966613,
      "grad_norm": 0.7029590606689453,
      "learning_rate": 3.103830936124775e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 27183040,
      "step": 16895
    },
    {
      "epoch": 0.765069376853256,
      "grad_norm": 0.5412904620170593,
      "learning_rate": 3.102878208724018e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 27190976,
      "step": 16900
    },
    {
      "epoch": 0.765295728739899,
      "grad_norm": 0.2830488979816437,
      "learning_rate": 3.101925388351865e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 27199968,
      "step": 16905
    },
    {
      "epoch": 0.765522080626542,
      "grad_norm": 0.6192075610160828,
      "learning_rate": 3.1009724751552515e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 27208448,
      "step": 16910
    },
    {
      "epoch": 0.765748432513185,
      "grad_norm": 0.8152726292610168,
      "learning_rate": 3.100019469281131e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 27216384,
      "step": 16915
    },
    {
      "epoch": 0.765974784399828,
      "grad_norm": 0.4905663728713989,
      "learning_rate": 3.0990663708764685e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 27224224,
      "step": 16920
    },
    {
      "epoch": 0.7662011362864709,
      "grad_norm": 0.546675443649292,
      "learning_rate": 3.098113180088243e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 27232064,
      "step": 16925
    },
    {
      "epoch": 0.7664274881731139,
      "grad_norm": 0.7076300978660583,
      "learning_rate": 3.097159897063448e-05,
      "loss": 0.1632,
      "num_input_tokens_seen": 27239520,
      "step": 16930
    },
    {
      "epoch": 0.7666538400597569,
      "grad_norm": 0.7169869542121887,
      "learning_rate": 3.096206521949094e-05,
      "loss": 0.1819,
      "num_input_tokens_seen": 27246976,
      "step": 16935
    },
    {
      "epoch": 0.7668801919463999,
      "grad_norm": 0.7784725427627563,
      "learning_rate": 3.0952530548922006e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 27255104,
      "step": 16940
    },
    {
      "epoch": 0.7671065438330429,
      "grad_norm": 0.22359804809093475,
      "learning_rate": 3.0942994960398064e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 27262848,
      "step": 16945
    },
    {
      "epoch": 0.7673328957196858,
      "grad_norm": 0.48677945137023926,
      "learning_rate": 3.093345845538961e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 27271520,
      "step": 16950
    },
    {
      "epoch": 0.7675592476063288,
      "grad_norm": 1.2398205995559692,
      "learning_rate": 3.09239210353673e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 27278880,
      "step": 16955
    },
    {
      "epoch": 0.7677855994929718,
      "grad_norm": 0.8024700880050659,
      "learning_rate": 3.0914382701801926e-05,
      "loss": 0.1083,
      "num_input_tokens_seen": 27286464,
      "step": 16960
    },
    {
      "epoch": 0.7680119513796148,
      "grad_norm": 0.4856206178665161,
      "learning_rate": 3.090484345616441e-05,
      "loss": 0.1175,
      "num_input_tokens_seen": 27294048,
      "step": 16965
    },
    {
      "epoch": 0.7682383032662577,
      "grad_norm": 0.29001525044441223,
      "learning_rate": 3.0895303299925825e-05,
      "loss": 0.1952,
      "num_input_tokens_seen": 27302624,
      "step": 16970
    },
    {
      "epoch": 0.7684646551529007,
      "grad_norm": 0.5958074927330017,
      "learning_rate": 3.0885762234557393e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 27310144,
      "step": 16975
    },
    {
      "epoch": 0.7686910070395436,
      "grad_norm": 0.7205127477645874,
      "learning_rate": 3.087622026153045e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 27318368,
      "step": 16980
    },
    {
      "epoch": 0.7689173589261866,
      "grad_norm": 0.41815581917762756,
      "learning_rate": 3.086667738231651e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 27326272,
      "step": 16985
    },
    {
      "epoch": 0.7691437108128296,
      "grad_norm": 1.3088792562484741,
      "learning_rate": 3.085713359838718e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 27334048,
      "step": 16990
    },
    {
      "epoch": 0.7693700626994726,
      "grad_norm": 0.33537760376930237,
      "learning_rate": 3.084758891121425e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 27342112,
      "step": 16995
    },
    {
      "epoch": 0.7695964145861156,
      "grad_norm": 0.44541463255882263,
      "learning_rate": 3.083804332226963e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 27349920,
      "step": 17000
    },
    {
      "epoch": 0.7695964145861156,
      "eval_loss": 0.1500329226255417,
      "eval_runtime": 403.6906,
      "eval_samples_per_second": 97.28,
      "eval_steps_per_second": 24.321,
      "num_input_tokens_seen": 27349920,
      "step": 17000
    },
    {
      "epoch": 0.7698227664727586,
      "grad_norm": 0.6574888229370117,
      "learning_rate": 3.082849683302536e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 27357728,
      "step": 17005
    },
    {
      "epoch": 0.7700491183594015,
      "grad_norm": 0.6460491418838501,
      "learning_rate": 3.081894944495363e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 27365568,
      "step": 17010
    },
    {
      "epoch": 0.7702754702460445,
      "grad_norm": 0.6155640482902527,
      "learning_rate": 3.080940115952677e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 27373696,
      "step": 17015
    },
    {
      "epoch": 0.7705018221326875,
      "grad_norm": 0.5156596899032593,
      "learning_rate": 3.0799851978217245e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 27381824,
      "step": 17020
    },
    {
      "epoch": 0.7707281740193305,
      "grad_norm": 0.4733457565307617,
      "learning_rate": 3.0790301902497666e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 27390304,
      "step": 17025
    },
    {
      "epoch": 0.7709545259059735,
      "grad_norm": 0.49879440665245056,
      "learning_rate": 3.078075093384076e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 27398592,
      "step": 17030
    },
    {
      "epoch": 0.7711808777926163,
      "grad_norm": 0.763724148273468,
      "learning_rate": 3.077119907371942e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 27406144,
      "step": 17035
    },
    {
      "epoch": 0.7714072296792593,
      "grad_norm": 0.9202253222465515,
      "learning_rate": 3.076164632360666e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 27413952,
      "step": 17040
    },
    {
      "epoch": 0.7716335815659023,
      "grad_norm": 0.662919282913208,
      "learning_rate": 3.075209268497563e-05,
      "loss": 0.1667,
      "num_input_tokens_seen": 27422048,
      "step": 17045
    },
    {
      "epoch": 0.7718599334525453,
      "grad_norm": 0.3487548530101776,
      "learning_rate": 3.074253815929961e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 27429536,
      "step": 17050
    },
    {
      "epoch": 0.7720862853391883,
      "grad_norm": 0.33891981840133667,
      "learning_rate": 3.0732982748052054e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 27438048,
      "step": 17055
    },
    {
      "epoch": 0.7723126372258313,
      "grad_norm": 0.8999518156051636,
      "learning_rate": 3.072342645270651e-05,
      "loss": 0.1206,
      "num_input_tokens_seen": 27445856,
      "step": 17060
    },
    {
      "epoch": 0.7725389891124742,
      "grad_norm": 0.7127499580383301,
      "learning_rate": 3.071386927473668e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 27454112,
      "step": 17065
    },
    {
      "epoch": 0.7727653409991172,
      "grad_norm": 0.4489235579967499,
      "learning_rate": 3.0704311215616404e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 27461824,
      "step": 17070
    },
    {
      "epoch": 0.7729916928857602,
      "grad_norm": 0.8808361887931824,
      "learning_rate": 3.0694752276819656e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 27469344,
      "step": 17075
    },
    {
      "epoch": 0.7732180447724032,
      "grad_norm": 0.3705165684223175,
      "learning_rate": 3.068519245982054e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 27477056,
      "step": 17080
    },
    {
      "epoch": 0.7734443966590462,
      "grad_norm": 0.8864201307296753,
      "learning_rate": 3.0675631766093304e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 27485632,
      "step": 17085
    },
    {
      "epoch": 0.7736707485456892,
      "grad_norm": 0.6740044951438904,
      "learning_rate": 3.066607019711232e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 27493824,
      "step": 17090
    },
    {
      "epoch": 0.7738971004323321,
      "grad_norm": 0.7267778515815735,
      "learning_rate": 3.065650775435211e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 27501984,
      "step": 17095
    },
    {
      "epoch": 0.7741234523189751,
      "grad_norm": 0.4636858105659485,
      "learning_rate": 3.0646944439287326e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 27509760,
      "step": 17100
    },
    {
      "epoch": 0.774349804205618,
      "grad_norm": 0.5677172541618347,
      "learning_rate": 3.0637380253392736e-05,
      "loss": 0.1812,
      "num_input_tokens_seen": 27517504,
      "step": 17105
    },
    {
      "epoch": 0.774576156092261,
      "grad_norm": 0.40335842967033386,
      "learning_rate": 3.062781519814327e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 27525920,
      "step": 17110
    },
    {
      "epoch": 0.774802507978904,
      "grad_norm": 0.6101309061050415,
      "learning_rate": 3.0618249275013985e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 27534464,
      "step": 17115
    },
    {
      "epoch": 0.7750288598655469,
      "grad_norm": 0.5912384986877441,
      "learning_rate": 3.060868248548005e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 27542368,
      "step": 17120
    },
    {
      "epoch": 0.7752552117521899,
      "grad_norm": 0.874983549118042,
      "learning_rate": 3.0599114831016796e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 27550176,
      "step": 17125
    },
    {
      "epoch": 0.7754815636388329,
      "grad_norm": 0.855372428894043,
      "learning_rate": 3.0589546313099666e-05,
      "loss": 0.1804,
      "num_input_tokens_seen": 27558592,
      "step": 17130
    },
    {
      "epoch": 0.7757079155254759,
      "grad_norm": 0.7488910555839539,
      "learning_rate": 3.0579976933204255e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 27566720,
      "step": 17135
    },
    {
      "epoch": 0.7759342674121189,
      "grad_norm": 0.7119935154914856,
      "learning_rate": 3.0570406692806284e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 27575296,
      "step": 17140
    },
    {
      "epoch": 0.7761606192987619,
      "grad_norm": 1.2822092771530151,
      "learning_rate": 3.05608355933816e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 27583072,
      "step": 17145
    },
    {
      "epoch": 0.7763869711854048,
      "grad_norm": 0.5312118530273438,
      "learning_rate": 3.055126363640618e-05,
      "loss": 0.1028,
      "num_input_tokens_seen": 27590816,
      "step": 17150
    },
    {
      "epoch": 0.7766133230720478,
      "grad_norm": 0.9711290001869202,
      "learning_rate": 3.0541690823356146e-05,
      "loss": 0.1209,
      "num_input_tokens_seen": 27598848,
      "step": 17155
    },
    {
      "epoch": 0.7768396749586908,
      "grad_norm": 0.3726673126220703,
      "learning_rate": 3.053211715570775e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 27606848,
      "step": 17160
    },
    {
      "epoch": 0.7770660268453338,
      "grad_norm": 0.3938332200050354,
      "learning_rate": 3.052254263493736e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 27614432,
      "step": 17165
    },
    {
      "epoch": 0.7772923787319768,
      "grad_norm": 1.2384129762649536,
      "learning_rate": 3.0512967262521498e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 27622400,
      "step": 17170
    },
    {
      "epoch": 0.7775187306186198,
      "grad_norm": 1.1731882095336914,
      "learning_rate": 3.0503391039936803e-05,
      "loss": 0.1757,
      "num_input_tokens_seen": 27630592,
      "step": 17175
    },
    {
      "epoch": 0.7777450825052626,
      "grad_norm": 0.3040612041950226,
      "learning_rate": 3.0493813968660056e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 27638752,
      "step": 17180
    },
    {
      "epoch": 0.7779714343919056,
      "grad_norm": 0.35477888584136963,
      "learning_rate": 3.0484236050168153e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 27646880,
      "step": 17185
    },
    {
      "epoch": 0.7781977862785486,
      "grad_norm": 0.8523713946342468,
      "learning_rate": 3.0474657285938123e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 27655360,
      "step": 17190
    },
    {
      "epoch": 0.7784241381651916,
      "grad_norm": 0.6532542705535889,
      "learning_rate": 3.046507767744715e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 27663008,
      "step": 17195
    },
    {
      "epoch": 0.7786504900518346,
      "grad_norm": 0.6487079858779907,
      "learning_rate": 3.045549722617252e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 27671008,
      "step": 17200
    },
    {
      "epoch": 0.7786504900518346,
      "eval_loss": 0.15048104524612427,
      "eval_runtime": 403.7283,
      "eval_samples_per_second": 97.271,
      "eval_steps_per_second": 24.318,
      "num_input_tokens_seen": 27671008,
      "step": 17200
    },
    {
      "epoch": 0.7788768419384775,
      "grad_norm": 0.8535112142562866,
      "learning_rate": 3.0445915933591658e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 27678560,
      "step": 17205
    },
    {
      "epoch": 0.7791031938251205,
      "grad_norm": 0.503977358341217,
      "learning_rate": 3.0436333801182114e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 27686592,
      "step": 17210
    },
    {
      "epoch": 0.7793295457117635,
      "grad_norm": 0.42131245136260986,
      "learning_rate": 3.0426750830421596e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 27694400,
      "step": 17215
    },
    {
      "epoch": 0.7795558975984065,
      "grad_norm": 1.1983270645141602,
      "learning_rate": 3.0417167022787897e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 27702944,
      "step": 17220
    },
    {
      "epoch": 0.7797822494850495,
      "grad_norm": 0.7849477529525757,
      "learning_rate": 3.0407582379758966e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 27711168,
      "step": 17225
    },
    {
      "epoch": 0.7800086013716925,
      "grad_norm": 1.0481928586959839,
      "learning_rate": 3.039799690281287e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 27719392,
      "step": 17230
    },
    {
      "epoch": 0.7802349532583354,
      "grad_norm": 0.6270567774772644,
      "learning_rate": 3.0388410593427823e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 27727776,
      "step": 17235
    },
    {
      "epoch": 0.7804613051449784,
      "grad_norm": 0.4156133830547333,
      "learning_rate": 3.0378823453082146e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 27736320,
      "step": 17240
    },
    {
      "epoch": 0.7806876570316214,
      "grad_norm": 0.8051946759223938,
      "learning_rate": 3.03692354832543e-05,
      "loss": 0.1872,
      "num_input_tokens_seen": 27744224,
      "step": 17245
    },
    {
      "epoch": 0.7809140089182643,
      "grad_norm": 0.7694629430770874,
      "learning_rate": 3.0359646685422865e-05,
      "loss": 0.1652,
      "num_input_tokens_seen": 27751328,
      "step": 17250
    },
    {
      "epoch": 0.7811403608049073,
      "grad_norm": 0.39106282591819763,
      "learning_rate": 3.035005706106656e-05,
      "loss": 0.1879,
      "num_input_tokens_seen": 27759264,
      "step": 17255
    },
    {
      "epoch": 0.7813667126915503,
      "grad_norm": 0.8280254602432251,
      "learning_rate": 3.034046661166422e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 27767296,
      "step": 17260
    },
    {
      "epoch": 0.7815930645781932,
      "grad_norm": 0.48700350522994995,
      "learning_rate": 3.033087533869482e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 27775104,
      "step": 17265
    },
    {
      "epoch": 0.7818194164648362,
      "grad_norm": 0.9624695181846619,
      "learning_rate": 3.0321283243637444e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 27782816,
      "step": 17270
    },
    {
      "epoch": 0.7820457683514792,
      "grad_norm": 0.44722041487693787,
      "learning_rate": 3.0311690327971326e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 27790400,
      "step": 17275
    },
    {
      "epoch": 0.7822721202381222,
      "grad_norm": 0.4279118478298187,
      "learning_rate": 3.030209659317581e-05,
      "loss": 0.1918,
      "num_input_tokens_seen": 27798368,
      "step": 17280
    },
    {
      "epoch": 0.7824984721247652,
      "grad_norm": 0.40081316232681274,
      "learning_rate": 3.0292502040730362e-05,
      "loss": 0.1534,
      "num_input_tokens_seen": 27807232,
      "step": 17285
    },
    {
      "epoch": 0.7827248240114081,
      "grad_norm": 0.6326753497123718,
      "learning_rate": 3.0282906672114597e-05,
      "loss": 0.1077,
      "num_input_tokens_seen": 27814816,
      "step": 17290
    },
    {
      "epoch": 0.7829511758980511,
      "grad_norm": 0.4177990257740021,
      "learning_rate": 3.027331048880823e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 27822976,
      "step": 17295
    },
    {
      "epoch": 0.7831775277846941,
      "grad_norm": 0.3982324004173279,
      "learning_rate": 3.0263713492291123e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 27830656,
      "step": 17300
    },
    {
      "epoch": 0.7834038796713371,
      "grad_norm": 1.6897609233856201,
      "learning_rate": 3.0254115684043242e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 27838368,
      "step": 17305
    },
    {
      "epoch": 0.7836302315579801,
      "grad_norm": 0.5011301040649414,
      "learning_rate": 3.024451706554469e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 27846880,
      "step": 17310
    },
    {
      "epoch": 0.7838565834446231,
      "grad_norm": 0.6548916697502136,
      "learning_rate": 3.0234917638275705e-05,
      "loss": 0.1225,
      "num_input_tokens_seen": 27854208,
      "step": 17315
    },
    {
      "epoch": 0.7840829353312659,
      "grad_norm": 0.4382435083389282,
      "learning_rate": 3.0225317403716635e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 27862208,
      "step": 17320
    },
    {
      "epoch": 0.7843092872179089,
      "grad_norm": 0.8598760962486267,
      "learning_rate": 3.0215716363347956e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 27871424,
      "step": 17325
    },
    {
      "epoch": 0.7845356391045519,
      "grad_norm": 0.6435909271240234,
      "learning_rate": 3.0206114518650275e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 27880544,
      "step": 17330
    },
    {
      "epoch": 0.7847619909911949,
      "grad_norm": 0.827903151512146,
      "learning_rate": 3.0196511871104304e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 27888352,
      "step": 17335
    },
    {
      "epoch": 0.7849883428778379,
      "grad_norm": 0.4860205054283142,
      "learning_rate": 3.01869084221909e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 27896000,
      "step": 17340
    },
    {
      "epoch": 0.7852146947644809,
      "grad_norm": 0.5817036628723145,
      "learning_rate": 3.0177304173391037e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 27904192,
      "step": 17345
    },
    {
      "epoch": 0.7854410466511238,
      "grad_norm": 0.6737185120582581,
      "learning_rate": 3.01676991261858e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 27912128,
      "step": 17350
    },
    {
      "epoch": 0.7856673985377668,
      "grad_norm": 0.5628127455711365,
      "learning_rate": 3.015809328205642e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 27920000,
      "step": 17355
    },
    {
      "epoch": 0.7858937504244098,
      "grad_norm": 0.6253569722175598,
      "learning_rate": 3.0148486642484248e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 27927968,
      "step": 17360
    },
    {
      "epoch": 0.7861201023110528,
      "grad_norm": 0.46705976128578186,
      "learning_rate": 3.0138879208950722e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 27936224,
      "step": 17365
    },
    {
      "epoch": 0.7863464541976958,
      "grad_norm": 1.4243850708007812,
      "learning_rate": 3.012927098293744e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 27944192,
      "step": 17370
    },
    {
      "epoch": 0.7865728060843387,
      "grad_norm": 0.7013789415359497,
      "learning_rate": 3.0119661965926123e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 27951904,
      "step": 17375
    },
    {
      "epoch": 0.7867991579709817,
      "grad_norm": 0.5510812401771545,
      "learning_rate": 3.0110052159398587e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 27960256,
      "step": 17380
    },
    {
      "epoch": 0.7870255098576246,
      "grad_norm": 0.5052339434623718,
      "learning_rate": 3.0100441564836802e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 27968512,
      "step": 17385
    },
    {
      "epoch": 0.7872518617442676,
      "grad_norm": 0.5376572608947754,
      "learning_rate": 3.0090830183722817e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 27977344,
      "step": 17390
    },
    {
      "epoch": 0.7874782136309106,
      "grad_norm": 0.6624816060066223,
      "learning_rate": 3.0081218017538852e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 27985216,
      "step": 17395
    },
    {
      "epoch": 0.7877045655175536,
      "grad_norm": 0.2671644389629364,
      "learning_rate": 3.0071605067767212e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 27993280,
      "step": 17400
    },
    {
      "epoch": 0.7877045655175536,
      "eval_loss": 0.14914661645889282,
      "eval_runtime": 404.2983,
      "eval_samples_per_second": 97.134,
      "eval_steps_per_second": 24.284,
      "num_input_tokens_seen": 27993280,
      "step": 17400
    },
    {
      "epoch": 0.7879309174041965,
      "grad_norm": 0.4911615252494812,
      "learning_rate": 3.006199133589034e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 28001312,
      "step": 17405
    },
    {
      "epoch": 0.7881572692908395,
      "grad_norm": 0.8207592964172363,
      "learning_rate": 3.005237682339079e-05,
      "loss": 0.0951,
      "num_input_tokens_seen": 28009184,
      "step": 17410
    },
    {
      "epoch": 0.7883836211774825,
      "grad_norm": 0.3037514090538025,
      "learning_rate": 3.0042761531751228e-05,
      "loss": 0.0898,
      "num_input_tokens_seen": 28017984,
      "step": 17415
    },
    {
      "epoch": 0.7886099730641255,
      "grad_norm": 0.6505535244941711,
      "learning_rate": 3.0033145462454482e-05,
      "loss": 0.1619,
      "num_input_tokens_seen": 28026080,
      "step": 17420
    },
    {
      "epoch": 0.7888363249507685,
      "grad_norm": 0.961919903755188,
      "learning_rate": 3.002352861698345e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 28033984,
      "step": 17425
    },
    {
      "epoch": 0.7890626768374115,
      "grad_norm": 1.0526214838027954,
      "learning_rate": 3.0013910996821178e-05,
      "loss": 0.1043,
      "num_input_tokens_seen": 28042528,
      "step": 17430
    },
    {
      "epoch": 0.7892890287240544,
      "grad_norm": 1.6532524824142456,
      "learning_rate": 3.0004292603450817e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 28050464,
      "step": 17435
    },
    {
      "epoch": 0.7895153806106974,
      "grad_norm": 0.873772382736206,
      "learning_rate": 2.9994673438355653e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 28058720,
      "step": 17440
    },
    {
      "epoch": 0.7897417324973404,
      "grad_norm": 0.8121710419654846,
      "learning_rate": 2.9985053503019078e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 28066688,
      "step": 17445
    },
    {
      "epoch": 0.7899680843839834,
      "grad_norm": 0.5637089014053345,
      "learning_rate": 2.99754327989246e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 28075360,
      "step": 17450
    },
    {
      "epoch": 0.7901944362706264,
      "grad_norm": 0.6017326712608337,
      "learning_rate": 2.9965811327555864e-05,
      "loss": 0.2027,
      "num_input_tokens_seen": 28083552,
      "step": 17455
    },
    {
      "epoch": 0.7904207881572692,
      "grad_norm": 0.8945175409317017,
      "learning_rate": 2.995618909039662e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 28091296,
      "step": 17460
    },
    {
      "epoch": 0.7906471400439122,
      "grad_norm": 0.6060025095939636,
      "learning_rate": 2.9946566088930727e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 28099136,
      "step": 17465
    },
    {
      "epoch": 0.7908734919305552,
      "grad_norm": 0.6573881506919861,
      "learning_rate": 2.9936942324642192e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 28107040,
      "step": 17470
    },
    {
      "epoch": 0.7910998438171982,
      "grad_norm": 0.6035355925559998,
      "learning_rate": 2.9927317799015097e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 28114784,
      "step": 17475
    },
    {
      "epoch": 0.7913261957038412,
      "grad_norm": 0.5335808992385864,
      "learning_rate": 2.9917692513533685e-05,
      "loss": 0.1195,
      "num_input_tokens_seen": 28122624,
      "step": 17480
    },
    {
      "epoch": 0.7915525475904842,
      "grad_norm": 1.1688066720962524,
      "learning_rate": 2.990806646968229e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 28130944,
      "step": 17485
    },
    {
      "epoch": 0.7917788994771271,
      "grad_norm": 0.8286226987838745,
      "learning_rate": 2.989843966894536e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 28138944,
      "step": 17490
    },
    {
      "epoch": 0.7920052513637701,
      "grad_norm": 0.7712663412094116,
      "learning_rate": 2.9888812112807472e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 28147264,
      "step": 17495
    },
    {
      "epoch": 0.7922316032504131,
      "grad_norm": 0.4365239143371582,
      "learning_rate": 2.987918380275333e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 28156000,
      "step": 17500
    },
    {
      "epoch": 0.7924579551370561,
      "grad_norm": 1.2191075086593628,
      "learning_rate": 2.9869554740267724e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 28163552,
      "step": 17505
    },
    {
      "epoch": 0.7926843070236991,
      "grad_norm": 0.3327077329158783,
      "learning_rate": 2.9859924926835585e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 28172416,
      "step": 17510
    },
    {
      "epoch": 0.7929106589103421,
      "grad_norm": 0.5600362420082092,
      "learning_rate": 2.9850294363941944e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 28181152,
      "step": 17515
    },
    {
      "epoch": 0.793137010796985,
      "grad_norm": 0.5296432375907898,
      "learning_rate": 2.9840663053071967e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 28189312,
      "step": 17520
    },
    {
      "epoch": 0.793363362683628,
      "grad_norm": 0.3330000638961792,
      "learning_rate": 2.983103099571091e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 28197312,
      "step": 17525
    },
    {
      "epoch": 0.7935897145702709,
      "grad_norm": 0.7273134589195251,
      "learning_rate": 2.9821398193344164e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 28206080,
      "step": 17530
    },
    {
      "epoch": 0.7938160664569139,
      "grad_norm": 0.3941895067691803,
      "learning_rate": 2.9811764647457226e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 28214048,
      "step": 17535
    },
    {
      "epoch": 0.7940424183435569,
      "grad_norm": 0.7628083825111389,
      "learning_rate": 2.9802130359535714e-05,
      "loss": 0.1956,
      "num_input_tokens_seen": 28222496,
      "step": 17540
    },
    {
      "epoch": 0.7942687702301998,
      "grad_norm": 0.6651185154914856,
      "learning_rate": 2.979249533106535e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 28230624,
      "step": 17545
    },
    {
      "epoch": 0.7944951221168428,
      "grad_norm": 0.37804314494132996,
      "learning_rate": 2.9782859563531986e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 28239008,
      "step": 17550
    },
    {
      "epoch": 0.7947214740034858,
      "grad_norm": 0.6877592206001282,
      "learning_rate": 2.977322305842156e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 28247520,
      "step": 17555
    },
    {
      "epoch": 0.7949478258901288,
      "grad_norm": 0.6016725897789001,
      "learning_rate": 2.9763585817220162e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 28255200,
      "step": 17560
    },
    {
      "epoch": 0.7951741777767718,
      "grad_norm": 0.4569206237792969,
      "learning_rate": 2.975394784141397e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 28264448,
      "step": 17565
    },
    {
      "epoch": 0.7954005296634148,
      "grad_norm": 1.2761025428771973,
      "learning_rate": 2.974430913248928e-05,
      "loss": 0.1839,
      "num_input_tokens_seen": 28272256,
      "step": 17570
    },
    {
      "epoch": 0.7956268815500577,
      "grad_norm": 0.2568283677101135,
      "learning_rate": 2.9734669691932497e-05,
      "loss": 0.1033,
      "num_input_tokens_seen": 28280224,
      "step": 17575
    },
    {
      "epoch": 0.7958532334367007,
      "grad_norm": 0.9274720549583435,
      "learning_rate": 2.9725029521230147e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 28288064,
      "step": 17580
    },
    {
      "epoch": 0.7960795853233437,
      "grad_norm": 0.9468487501144409,
      "learning_rate": 2.9715388621868873e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 28296096,
      "step": 17585
    },
    {
      "epoch": 0.7963059372099867,
      "grad_norm": 0.9202572703361511,
      "learning_rate": 2.970574699533541e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 28304096,
      "step": 17590
    },
    {
      "epoch": 0.7965322890966297,
      "grad_norm": 0.368026465177536,
      "learning_rate": 2.969610464311662e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 28311680,
      "step": 17595
    },
    {
      "epoch": 0.7967586409832726,
      "grad_norm": 1.52545166015625,
      "learning_rate": 2.9686461566699487e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 28319712,
      "step": 17600
    },
    {
      "epoch": 0.7967586409832726,
      "eval_loss": 0.1493002474308014,
      "eval_runtime": 404.1773,
      "eval_samples_per_second": 97.163,
      "eval_steps_per_second": 24.291,
      "num_input_tokens_seen": 28319712,
      "step": 17600
    },
    {
      "epoch": 0.7969849928699155,
      "grad_norm": 0.49765822291374207,
      "learning_rate": 2.9676817767571086e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 28327552,
      "step": 17605
    },
    {
      "epoch": 0.7972113447565585,
      "grad_norm": 0.37891489267349243,
      "learning_rate": 2.966717324721861e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 28335232,
      "step": 17610
    },
    {
      "epoch": 0.7974376966432015,
      "grad_norm": 0.4097761809825897,
      "learning_rate": 2.9657528007129366e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 28344256,
      "step": 17615
    },
    {
      "epoch": 0.7976640485298445,
      "grad_norm": 0.48573926091194153,
      "learning_rate": 2.9647882048790777e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 28351968,
      "step": 17620
    },
    {
      "epoch": 0.7978904004164875,
      "grad_norm": 0.9598826169967651,
      "learning_rate": 2.963823537369037e-05,
      "loss": 0.0982,
      "num_input_tokens_seen": 28360224,
      "step": 17625
    },
    {
      "epoch": 0.7981167523031304,
      "grad_norm": 0.5791976451873779,
      "learning_rate": 2.9628587983315775e-05,
      "loss": 0.1788,
      "num_input_tokens_seen": 28367616,
      "step": 17630
    },
    {
      "epoch": 0.7983431041897734,
      "grad_norm": 0.39175137877464294,
      "learning_rate": 2.9618939879154746e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 28376544,
      "step": 17635
    },
    {
      "epoch": 0.7985694560764164,
      "grad_norm": 1.0258772373199463,
      "learning_rate": 2.9609291062695143e-05,
      "loss": 0.1939,
      "num_input_tokens_seen": 28384480,
      "step": 17640
    },
    {
      "epoch": 0.7987958079630594,
      "grad_norm": 0.39928939938545227,
      "learning_rate": 2.9599641535424938e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 28392096,
      "step": 17645
    },
    {
      "epoch": 0.7990221598497024,
      "grad_norm": 0.9726661443710327,
      "learning_rate": 2.9589991298832202e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 28400416,
      "step": 17650
    },
    {
      "epoch": 0.7992485117363454,
      "grad_norm": 0.6083806753158569,
      "learning_rate": 2.958034035440513e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 28407808,
      "step": 17655
    },
    {
      "epoch": 0.7994748636229883,
      "grad_norm": 0.7990466952323914,
      "learning_rate": 2.957068870363201e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 28415872,
      "step": 17660
    },
    {
      "epoch": 0.7997012155096312,
      "grad_norm": 0.47416356205940247,
      "learning_rate": 2.956103634800126e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 28423776,
      "step": 17665
    },
    {
      "epoch": 0.7999275673962742,
      "grad_norm": 0.5595195293426514,
      "learning_rate": 2.9551383289001384e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 28431328,
      "step": 17670
    },
    {
      "epoch": 0.8001539192829172,
      "grad_norm": 0.3948931097984314,
      "learning_rate": 2.9541729528121005e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 28439520,
      "step": 17675
    },
    {
      "epoch": 0.8003802711695602,
      "grad_norm": 0.6742562055587769,
      "learning_rate": 2.9532075066848856e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 28447456,
      "step": 17680
    },
    {
      "epoch": 0.8006066230562032,
      "grad_norm": 0.6923664212226868,
      "learning_rate": 2.9522419906673786e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 28455200,
      "step": 17685
    },
    {
      "epoch": 0.8008329749428461,
      "grad_norm": 0.7573985457420349,
      "learning_rate": 2.951276404908474e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 28463616,
      "step": 17690
    },
    {
      "epoch": 0.8010593268294891,
      "grad_norm": 0.5578569769859314,
      "learning_rate": 2.9503107495570752e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 28471040,
      "step": 17695
    },
    {
      "epoch": 0.8012856787161321,
      "grad_norm": 0.8692209720611572,
      "learning_rate": 2.9493450247621003e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 28479200,
      "step": 17700
    },
    {
      "epoch": 0.8015120306027751,
      "grad_norm": 0.3885243535041809,
      "learning_rate": 2.948379230672476e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 28487296,
      "step": 17705
    },
    {
      "epoch": 0.8017383824894181,
      "grad_norm": 0.9225525856018066,
      "learning_rate": 2.9474133674371396e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 28496032,
      "step": 17710
    },
    {
      "epoch": 0.801964734376061,
      "grad_norm": 0.8459308743476868,
      "learning_rate": 2.9464474352050387e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 28504096,
      "step": 17715
    },
    {
      "epoch": 0.802191086262704,
      "grad_norm": 0.33946147561073303,
      "learning_rate": 2.9454814341251336e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 28512096,
      "step": 17720
    },
    {
      "epoch": 0.802417438149347,
      "grad_norm": 0.3863644301891327,
      "learning_rate": 2.9445153643463942e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 28519872,
      "step": 17725
    },
    {
      "epoch": 0.80264379003599,
      "grad_norm": 0.8573122620582581,
      "learning_rate": 2.943549226017798e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 28527936,
      "step": 17730
    },
    {
      "epoch": 0.802870141922633,
      "grad_norm": 0.7326724529266357,
      "learning_rate": 2.942583019288337e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 28536288,
      "step": 17735
    },
    {
      "epoch": 0.803096493809276,
      "grad_norm": 0.3192584812641144,
      "learning_rate": 2.9416167443070132e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 28544480,
      "step": 17740
    },
    {
      "epoch": 0.8033228456959188,
      "grad_norm": 1.030752420425415,
      "learning_rate": 2.9406504012228375e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 28552928,
      "step": 17745
    },
    {
      "epoch": 0.8035491975825618,
      "grad_norm": 0.33593353629112244,
      "learning_rate": 2.939683990184832e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 28560768,
      "step": 17750
    },
    {
      "epoch": 0.8037755494692048,
      "grad_norm": 0.7823719382286072,
      "learning_rate": 2.93871751134203e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 28568992,
      "step": 17755
    },
    {
      "epoch": 0.8040019013558478,
      "grad_norm": 1.5238010883331299,
      "learning_rate": 2.9377509648434752e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 28577280,
      "step": 17760
    },
    {
      "epoch": 0.8042282532424908,
      "grad_norm": 0.6584358215332031,
      "learning_rate": 2.9367843508382203e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 28584864,
      "step": 17765
    },
    {
      "epoch": 0.8044546051291338,
      "grad_norm": 0.4674878418445587,
      "learning_rate": 2.9358176694753293e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 28592896,
      "step": 17770
    },
    {
      "epoch": 0.8046809570157767,
      "grad_norm": 0.7966315150260925,
      "learning_rate": 2.9348509209038766e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 28601312,
      "step": 17775
    },
    {
      "epoch": 0.8049073089024197,
      "grad_norm": 0.39202094078063965,
      "learning_rate": 2.933884105272947e-05,
      "loss": 0.1664,
      "num_input_tokens_seen": 28609440,
      "step": 17780
    },
    {
      "epoch": 0.8051336607890627,
      "grad_norm": 0.5768570899963379,
      "learning_rate": 2.9329172227316366e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 28617280,
      "step": 17785
    },
    {
      "epoch": 0.8053600126757057,
      "grad_norm": 0.28052857518196106,
      "learning_rate": 2.93195027342905e-05,
      "loss": 0.1269,
      "num_input_tokens_seen": 28625216,
      "step": 17790
    },
    {
      "epoch": 0.8055863645623487,
      "grad_norm": 0.5038371086120605,
      "learning_rate": 2.9309832575143024e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 28633088,
      "step": 17795
    },
    {
      "epoch": 0.8058127164489916,
      "grad_norm": 0.582123339176178,
      "learning_rate": 2.930016175136521e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 28640832,
      "step": 17800
    },
    {
      "epoch": 0.8058127164489916,
      "eval_loss": 0.1488395631313324,
      "eval_runtime": 404.3751,
      "eval_samples_per_second": 97.115,
      "eval_steps_per_second": 24.279,
      "num_input_tokens_seen": 28640832,
      "step": 17800
    },
    {
      "epoch": 0.8060390683356345,
      "grad_norm": 0.5027723908424377,
      "learning_rate": 2.9290490264448412e-05,
      "loss": 0.1848,
      "num_input_tokens_seen": 28648608,
      "step": 17805
    },
    {
      "epoch": 0.8062654202222775,
      "grad_norm": 0.38300004601478577,
      "learning_rate": 2.9280818115884094e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 28656352,
      "step": 17810
    },
    {
      "epoch": 0.8064917721089205,
      "grad_norm": 0.7714638710021973,
      "learning_rate": 2.9271145307163828e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 28664096,
      "step": 17815
    },
    {
      "epoch": 0.8067181239955635,
      "grad_norm": 0.8470448851585388,
      "learning_rate": 2.9261471839779287e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 28671616,
      "step": 17820
    },
    {
      "epoch": 0.8069444758822065,
      "grad_norm": 0.4325641989707947,
      "learning_rate": 2.925179771522223e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 28680000,
      "step": 17825
    },
    {
      "epoch": 0.8071708277688494,
      "grad_norm": 0.388249933719635,
      "learning_rate": 2.9242122934984535e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 28688160,
      "step": 17830
    },
    {
      "epoch": 0.8073971796554924,
      "grad_norm": 0.589012086391449,
      "learning_rate": 2.9232447500558176e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 28696608,
      "step": 17835
    },
    {
      "epoch": 0.8076235315421354,
      "grad_norm": 0.6139623522758484,
      "learning_rate": 2.9222771413435225e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 28705472,
      "step": 17840
    },
    {
      "epoch": 0.8078498834287784,
      "grad_norm": 0.8286152482032776,
      "learning_rate": 2.9213094675107848e-05,
      "loss": 0.159,
      "num_input_tokens_seen": 28713824,
      "step": 17845
    },
    {
      "epoch": 0.8080762353154214,
      "grad_norm": 1.0871750116348267,
      "learning_rate": 2.9203417287068335e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 28722560,
      "step": 17850
    },
    {
      "epoch": 0.8083025872020644,
      "grad_norm": 0.4257798194885254,
      "learning_rate": 2.9193739250809042e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 28730272,
      "step": 17855
    },
    {
      "epoch": 0.8085289390887073,
      "grad_norm": 1.1210145950317383,
      "learning_rate": 2.9184060567822463e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 28737664,
      "step": 17860
    },
    {
      "epoch": 0.8087552909753503,
      "grad_norm": 1.1468921899795532,
      "learning_rate": 2.9174381239601166e-05,
      "loss": 0.1729,
      "num_input_tokens_seen": 28745696,
      "step": 17865
    },
    {
      "epoch": 0.8089816428619933,
      "grad_norm": 0.9805024862289429,
      "learning_rate": 2.916470126763783e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 28754688,
      "step": 17870
    },
    {
      "epoch": 0.8092079947486363,
      "grad_norm": 0.825764000415802,
      "learning_rate": 2.9155020653425203e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 28763104,
      "step": 17875
    },
    {
      "epoch": 0.8094343466352792,
      "grad_norm": 0.7018896341323853,
      "learning_rate": 2.9145339398456184e-05,
      "loss": 0.1739,
      "num_input_tokens_seen": 28771680,
      "step": 17880
    },
    {
      "epoch": 0.8096606985219221,
      "grad_norm": 0.4396149516105652,
      "learning_rate": 2.913565750422374e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 28779744,
      "step": 17885
    },
    {
      "epoch": 0.8098870504085651,
      "grad_norm": 0.6040241718292236,
      "learning_rate": 2.9125974972220938e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 28788096,
      "step": 17890
    },
    {
      "epoch": 0.8101134022952081,
      "grad_norm": 0.30065715312957764,
      "learning_rate": 2.9116291803940932e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 28796064,
      "step": 17895
    },
    {
      "epoch": 0.8103397541818511,
      "grad_norm": 0.7379510402679443,
      "learning_rate": 2.910660800087701e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 28804480,
      "step": 17900
    },
    {
      "epoch": 0.8105661060684941,
      "grad_norm": 1.0556057691574097,
      "learning_rate": 2.909692356452254e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 28812480,
      "step": 17905
    },
    {
      "epoch": 0.8107924579551371,
      "grad_norm": 1.013287901878357,
      "learning_rate": 2.9087238496370962e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 28820416,
      "step": 17910
    },
    {
      "epoch": 0.81101880984178,
      "grad_norm": 0.7091326713562012,
      "learning_rate": 2.907755279791583e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 28828224,
      "step": 17915
    },
    {
      "epoch": 0.811245161728423,
      "grad_norm": 1.2085286378860474,
      "learning_rate": 2.906786647065083e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 28836160,
      "step": 17920
    },
    {
      "epoch": 0.811471513615066,
      "grad_norm": 0.3488149344921112,
      "learning_rate": 2.9058179516069695e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 28843776,
      "step": 17925
    },
    {
      "epoch": 0.811697865501709,
      "grad_norm": 0.6599149107933044,
      "learning_rate": 2.9048491935666282e-05,
      "loss": 0.1742,
      "num_input_tokens_seen": 28851648,
      "step": 17930
    },
    {
      "epoch": 0.811924217388352,
      "grad_norm": 1.0466355085372925,
      "learning_rate": 2.9038803730934534e-05,
      "loss": 0.1553,
      "num_input_tokens_seen": 28859712,
      "step": 17935
    },
    {
      "epoch": 0.812150569274995,
      "grad_norm": 0.6176345348358154,
      "learning_rate": 2.9029114903368503e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 28867584,
      "step": 17940
    },
    {
      "epoch": 0.8123769211616378,
      "grad_norm": 0.4594126045703888,
      "learning_rate": 2.9019425454462318e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 28874976,
      "step": 17945
    },
    {
      "epoch": 0.8126032730482808,
      "grad_norm": 0.32875967025756836,
      "learning_rate": 2.9009735385710212e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 28882880,
      "step": 17950
    },
    {
      "epoch": 0.8128296249349238,
      "grad_norm": 0.3707433044910431,
      "learning_rate": 2.900004469860652e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 28890688,
      "step": 17955
    },
    {
      "epoch": 0.8130559768215668,
      "grad_norm": 0.37953805923461914,
      "learning_rate": 2.8990353394645668e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 28898912,
      "step": 17960
    },
    {
      "epoch": 0.8132823287082098,
      "grad_norm": 0.5371087193489075,
      "learning_rate": 2.8980661475322186e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 28906752,
      "step": 17965
    },
    {
      "epoch": 0.8135086805948527,
      "grad_norm": 0.9348248243331909,
      "learning_rate": 2.897096894213067e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 28914272,
      "step": 17970
    },
    {
      "epoch": 0.8137350324814957,
      "grad_norm": 0.5497052073478699,
      "learning_rate": 2.8961275796565845e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 28922624,
      "step": 17975
    },
    {
      "epoch": 0.8139613843681387,
      "grad_norm": 0.3414757549762726,
      "learning_rate": 2.8951582040122517e-05,
      "loss": 0.1654,
      "num_input_tokens_seen": 28930880,
      "step": 17980
    },
    {
      "epoch": 0.8141877362547817,
      "grad_norm": 0.2995593547821045,
      "learning_rate": 2.894188767429557e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 28939072,
      "step": 17985
    },
    {
      "epoch": 0.8144140881414247,
      "grad_norm": 0.4899599552154541,
      "learning_rate": 2.8932192700580014e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 28946912,
      "step": 17990
    },
    {
      "epoch": 0.8146404400280677,
      "grad_norm": 0.7231727242469788,
      "learning_rate": 2.8922497120470916e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 28956096,
      "step": 17995
    },
    {
      "epoch": 0.8148667919147106,
      "grad_norm": 1.0998311042785645,
      "learning_rate": 2.891280093546348e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 28964096,
      "step": 18000
    },
    {
      "epoch": 0.8148667919147106,
      "eval_loss": 0.15039582550525665,
      "eval_runtime": 404.8434,
      "eval_samples_per_second": 97.003,
      "eval_steps_per_second": 24.251,
      "num_input_tokens_seen": 28964096,
      "step": 18000
    },
    {
      "epoch": 0.8150931438013536,
      "grad_norm": 0.5088233947753906,
      "learning_rate": 2.890310414705297e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 28972576,
      "step": 18005
    },
    {
      "epoch": 0.8153194956879966,
      "grad_norm": 0.9255858659744263,
      "learning_rate": 2.8893406756734742e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 28980288,
      "step": 18010
    },
    {
      "epoch": 0.8155458475746395,
      "grad_norm": 0.41820183396339417,
      "learning_rate": 2.888370876600427e-05,
      "loss": 0.1825,
      "num_input_tokens_seen": 28987968,
      "step": 18015
    },
    {
      "epoch": 0.8157721994612825,
      "grad_norm": 0.5590012073516846,
      "learning_rate": 2.8874010176357104e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 28996096,
      "step": 18020
    },
    {
      "epoch": 0.8159985513479255,
      "grad_norm": 0.9339338541030884,
      "learning_rate": 2.886431098928888e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 29004352,
      "step": 18025
    },
    {
      "epoch": 0.8162249032345684,
      "grad_norm": 0.7349648475646973,
      "learning_rate": 2.885461120629534e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 29012768,
      "step": 18030
    },
    {
      "epoch": 0.8164512551212114,
      "grad_norm": 0.9234617948532104,
      "learning_rate": 2.8844910828872317e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 29021312,
      "step": 18035
    },
    {
      "epoch": 0.8166776070078544,
      "grad_norm": 0.3124361038208008,
      "learning_rate": 2.8835209858515715e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 29029344,
      "step": 18040
    },
    {
      "epoch": 0.8169039588944974,
      "grad_norm": 0.3816421329975128,
      "learning_rate": 2.8825508296721566e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 29037376,
      "step": 18045
    },
    {
      "epoch": 0.8171303107811404,
      "grad_norm": 0.6590698957443237,
      "learning_rate": 2.881580614498596e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 29045568,
      "step": 18050
    },
    {
      "epoch": 0.8173566626677833,
      "grad_norm": 0.6372536420822144,
      "learning_rate": 2.8806103404805103e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 29053632,
      "step": 18055
    },
    {
      "epoch": 0.8175830145544263,
      "grad_norm": 0.45332860946655273,
      "learning_rate": 2.8796400077675257e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 29061536,
      "step": 18060
    },
    {
      "epoch": 0.8178093664410693,
      "grad_norm": 0.7815828919410706,
      "learning_rate": 2.8786696165092812e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 29070112,
      "step": 18065
    },
    {
      "epoch": 0.8180357183277123,
      "grad_norm": 0.5743731260299683,
      "learning_rate": 2.8776991668554236e-05,
      "loss": 0.1823,
      "num_input_tokens_seen": 29077792,
      "step": 18070
    },
    {
      "epoch": 0.8182620702143553,
      "grad_norm": 0.576393187046051,
      "learning_rate": 2.876728658955608e-05,
      "loss": 0.1258,
      "num_input_tokens_seen": 29085472,
      "step": 18075
    },
    {
      "epoch": 0.8184884221009983,
      "grad_norm": 0.2516724169254303,
      "learning_rate": 2.8757580929594986e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 29093536,
      "step": 18080
    },
    {
      "epoch": 0.8187147739876411,
      "grad_norm": 0.7524312138557434,
      "learning_rate": 2.87478746901677e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 29102560,
      "step": 18085
    },
    {
      "epoch": 0.8189411258742841,
      "grad_norm": 1.1260221004486084,
      "learning_rate": 2.873816787277103e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 29110496,
      "step": 18090
    },
    {
      "epoch": 0.8191674777609271,
      "grad_norm": 0.7039276957511902,
      "learning_rate": 2.8728460478901903e-05,
      "loss": 0.1791,
      "num_input_tokens_seen": 29118112,
      "step": 18095
    },
    {
      "epoch": 0.8193938296475701,
      "grad_norm": 0.8031821846961975,
      "learning_rate": 2.8718752510057307e-05,
      "loss": 0.1603,
      "num_input_tokens_seen": 29125568,
      "step": 18100
    },
    {
      "epoch": 0.8196201815342131,
      "grad_norm": 0.528891384601593,
      "learning_rate": 2.870904396773435e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 29134080,
      "step": 18105
    },
    {
      "epoch": 0.8198465334208561,
      "grad_norm": 0.4738764762878418,
      "learning_rate": 2.86993348534302e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 29142656,
      "step": 18110
    },
    {
      "epoch": 0.820072885307499,
      "grad_norm": 0.281968891620636,
      "learning_rate": 2.868962516864212e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 29150592,
      "step": 18115
    },
    {
      "epoch": 0.820299237194142,
      "grad_norm": 0.9581308960914612,
      "learning_rate": 2.8679914914867477e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 29159456,
      "step": 18120
    },
    {
      "epoch": 0.820525589080785,
      "grad_norm": 0.8024723529815674,
      "learning_rate": 2.8670204093603713e-05,
      "loss": 0.1223,
      "num_input_tokens_seen": 29167776,
      "step": 18125
    },
    {
      "epoch": 0.820751940967428,
      "grad_norm": 0.3868572413921356,
      "learning_rate": 2.8660492706348357e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 29175648,
      "step": 18130
    },
    {
      "epoch": 0.820978292854071,
      "grad_norm": 1.045061707496643,
      "learning_rate": 2.8650780754599022e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 29183584,
      "step": 18135
    },
    {
      "epoch": 0.8212046447407139,
      "grad_norm": 1.2895108461380005,
      "learning_rate": 2.8641068239853407e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 29191552,
      "step": 18140
    },
    {
      "epoch": 0.8214309966273569,
      "grad_norm": 0.8991073966026306,
      "learning_rate": 2.863135516360932e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 29199488,
      "step": 18145
    },
    {
      "epoch": 0.8216573485139999,
      "grad_norm": 0.9565503597259521,
      "learning_rate": 2.8621641527364633e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 29207296,
      "step": 18150
    },
    {
      "epoch": 0.8218837004006428,
      "grad_norm": 0.636172354221344,
      "learning_rate": 2.8611927332617313e-05,
      "loss": 0.1275,
      "num_input_tokens_seen": 29215072,
      "step": 18155
    },
    {
      "epoch": 0.8221100522872858,
      "grad_norm": 0.6991238594055176,
      "learning_rate": 2.8602212580865405e-05,
      "loss": 0.1689,
      "num_input_tokens_seen": 29223264,
      "step": 18160
    },
    {
      "epoch": 0.8223364041739288,
      "grad_norm": 0.6044691205024719,
      "learning_rate": 2.859249727360705e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 29230976,
      "step": 18165
    },
    {
      "epoch": 0.8225627560605717,
      "grad_norm": 0.8333984613418579,
      "learning_rate": 2.8582781412340465e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 29238368,
      "step": 18170
    },
    {
      "epoch": 0.8227891079472147,
      "grad_norm": 0.880501389503479,
      "learning_rate": 2.857306499856397e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 29246080,
      "step": 18175
    },
    {
      "epoch": 0.8230154598338577,
      "grad_norm": 0.4862043559551239,
      "learning_rate": 2.856334803377594e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 29254720,
      "step": 18180
    },
    {
      "epoch": 0.8232418117205007,
      "grad_norm": 0.648750901222229,
      "learning_rate": 2.8553630519474867e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 29262592,
      "step": 18185
    },
    {
      "epoch": 0.8234681636071437,
      "grad_norm": 0.2182074934244156,
      "learning_rate": 2.8543912457159317e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 29270304,
      "step": 18190
    },
    {
      "epoch": 0.8236945154937867,
      "grad_norm": 0.4155593514442444,
      "learning_rate": 2.853419384832792e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 29278432,
      "step": 18195
    },
    {
      "epoch": 0.8239208673804296,
      "grad_norm": 0.3396293818950653,
      "learning_rate": 2.8524474694479423e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 29286752,
      "step": 18200
    },
    {
      "epoch": 0.8239208673804296,
      "eval_loss": 0.14865413308143616,
      "eval_runtime": 404.6327,
      "eval_samples_per_second": 97.053,
      "eval_steps_per_second": 24.264,
      "num_input_tokens_seen": 29286752,
      "step": 18200
    },
    {
      "epoch": 0.8241472192670726,
      "grad_norm": 1.2220640182495117,
      "learning_rate": 2.851475499711264e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 29294528,
      "step": 18205
    },
    {
      "epoch": 0.8243735711537156,
      "grad_norm": 0.9617931842803955,
      "learning_rate": 2.8505034757726468e-05,
      "loss": 0.2045,
      "num_input_tokens_seen": 29302272,
      "step": 18210
    },
    {
      "epoch": 0.8245999230403586,
      "grad_norm": 0.4191884994506836,
      "learning_rate": 2.8495313977819886e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 29310656,
      "step": 18215
    },
    {
      "epoch": 0.8248262749270016,
      "grad_norm": 0.708336591720581,
      "learning_rate": 2.8485592658891956e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 29318848,
      "step": 18220
    },
    {
      "epoch": 0.8250526268136444,
      "grad_norm": 1.076654314994812,
      "learning_rate": 2.8475870802441844e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 29326048,
      "step": 18225
    },
    {
      "epoch": 0.8252789787002874,
      "grad_norm": 0.5464984774589539,
      "learning_rate": 2.8466148409968774e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 29333952,
      "step": 18230
    },
    {
      "epoch": 0.8255053305869304,
      "grad_norm": 0.4493934214115143,
      "learning_rate": 2.8456425482972067e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 29341536,
      "step": 18235
    },
    {
      "epoch": 0.8257316824735734,
      "grad_norm": 0.7701631784439087,
      "learning_rate": 2.84467020229511e-05,
      "loss": 0.1792,
      "num_input_tokens_seen": 29349472,
      "step": 18240
    },
    {
      "epoch": 0.8259580343602164,
      "grad_norm": 0.5637766718864441,
      "learning_rate": 2.8436978031405375e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 29357632,
      "step": 18245
    },
    {
      "epoch": 0.8261843862468594,
      "grad_norm": 0.656754732131958,
      "learning_rate": 2.842725350983445e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 29365952,
      "step": 18250
    },
    {
      "epoch": 0.8264107381335023,
      "grad_norm": 1.2670838832855225,
      "learning_rate": 2.8417528459737957e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 29374112,
      "step": 18255
    },
    {
      "epoch": 0.8266370900201453,
      "grad_norm": 1.430149793624878,
      "learning_rate": 2.8407802882615624e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 29381952,
      "step": 18260
    },
    {
      "epoch": 0.8268634419067883,
      "grad_norm": 0.9105744361877441,
      "learning_rate": 2.8398076779967277e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 29390240,
      "step": 18265
    },
    {
      "epoch": 0.8270897937934313,
      "grad_norm": 0.3018416166305542,
      "learning_rate": 2.8388350153292774e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 29397760,
      "step": 18270
    },
    {
      "epoch": 0.8273161456800743,
      "grad_norm": 1.547524333000183,
      "learning_rate": 2.8378623004092103e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 29405728,
      "step": 18275
    },
    {
      "epoch": 0.8275424975667173,
      "grad_norm": 0.9015833139419556,
      "learning_rate": 2.8368895333865302e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 29413472,
      "step": 18280
    },
    {
      "epoch": 0.8277688494533602,
      "grad_norm": 0.46103665232658386,
      "learning_rate": 2.835916714411251e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 29421376,
      "step": 18285
    },
    {
      "epoch": 0.8279952013400032,
      "grad_norm": 0.936835527420044,
      "learning_rate": 2.8349438436333926e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 29429152,
      "step": 18290
    },
    {
      "epoch": 0.8282215532266461,
      "grad_norm": 0.7632039189338684,
      "learning_rate": 2.833970921202984e-05,
      "loss": 0.1562,
      "num_input_tokens_seen": 29436832,
      "step": 18295
    },
    {
      "epoch": 0.8284479051132891,
      "grad_norm": 0.7222744226455688,
      "learning_rate": 2.8329979472700628e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 29445472,
      "step": 18300
    },
    {
      "epoch": 0.8286742569999321,
      "grad_norm": 1.6188684701919556,
      "learning_rate": 2.832024921984674e-05,
      "loss": 0.1128,
      "num_input_tokens_seen": 29454048,
      "step": 18305
    },
    {
      "epoch": 0.828900608886575,
      "grad_norm": 0.36522024869918823,
      "learning_rate": 2.8310518454968693e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 29462336,
      "step": 18310
    },
    {
      "epoch": 0.829126960773218,
      "grad_norm": 0.5001591444015503,
      "learning_rate": 2.8300787179567095e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 29470528,
      "step": 18315
    },
    {
      "epoch": 0.829353312659861,
      "grad_norm": 0.5712452530860901,
      "learning_rate": 2.8291055395142636e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 29478912,
      "step": 18320
    },
    {
      "epoch": 0.829579664546504,
      "grad_norm": 0.6044998168945312,
      "learning_rate": 2.8281323103196073e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 29487392,
      "step": 18325
    },
    {
      "epoch": 0.829806016433147,
      "grad_norm": 0.3571649491786957,
      "learning_rate": 2.8271590305228256e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 29494912,
      "step": 18330
    },
    {
      "epoch": 0.83003236831979,
      "grad_norm": 0.6384074687957764,
      "learning_rate": 2.82618570027401e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 29502976,
      "step": 18335
    },
    {
      "epoch": 0.8302587202064329,
      "grad_norm": 0.5641158223152161,
      "learning_rate": 2.8252123197232604e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 29510464,
      "step": 18340
    },
    {
      "epoch": 0.8304850720930759,
      "grad_norm": 0.5805977582931519,
      "learning_rate": 2.8242388890206843e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 29518464,
      "step": 18345
    },
    {
      "epoch": 0.8307114239797189,
      "grad_norm": 0.5205549001693726,
      "learning_rate": 2.8232654083163967e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 29526880,
      "step": 18350
    },
    {
      "epoch": 0.8309377758663619,
      "grad_norm": 0.7880983352661133,
      "learning_rate": 2.822291877760521e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 29534912,
      "step": 18355
    },
    {
      "epoch": 0.8311641277530049,
      "grad_norm": 0.743883490562439,
      "learning_rate": 2.8213182975031864e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 29543200,
      "step": 18360
    },
    {
      "epoch": 0.8313904796396479,
      "grad_norm": 0.6066312193870544,
      "learning_rate": 2.8203446676945337e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 29551488,
      "step": 18365
    },
    {
      "epoch": 0.8316168315262907,
      "grad_norm": 0.8010314702987671,
      "learning_rate": 2.8193709884847075e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 29559776,
      "step": 18370
    },
    {
      "epoch": 0.8318431834129337,
      "grad_norm": 1.3102210760116577,
      "learning_rate": 2.8183972600238605e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 29567168,
      "step": 18375
    },
    {
      "epoch": 0.8320695352995767,
      "grad_norm": 0.7361204028129578,
      "learning_rate": 2.817423482462156e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 29574976,
      "step": 18380
    },
    {
      "epoch": 0.8322958871862197,
      "grad_norm": 0.7124533653259277,
      "learning_rate": 2.8164496559497605e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 29582336,
      "step": 18385
    },
    {
      "epoch": 0.8325222390728627,
      "grad_norm": 0.5314837098121643,
      "learning_rate": 2.815475780636852e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 29590688,
      "step": 18390
    },
    {
      "epoch": 0.8327485909595056,
      "grad_norm": 0.5834685564041138,
      "learning_rate": 2.814501856673613e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 29598528,
      "step": 18395
    },
    {
      "epoch": 0.8329749428461486,
      "grad_norm": 0.5608265399932861,
      "learning_rate": 2.8135278842102353e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 29606528,
      "step": 18400
    },
    {
      "epoch": 0.8329749428461486,
      "eval_loss": 0.14934958517551422,
      "eval_runtime": 405.2353,
      "eval_samples_per_second": 96.909,
      "eval_steps_per_second": 24.228,
      "num_input_tokens_seen": 29606528,
      "step": 18400
    },
    {
      "epoch": 0.8332012947327916,
      "grad_norm": 0.33536046743392944,
      "learning_rate": 2.8125538633969183e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 29614880,
      "step": 18405
    },
    {
      "epoch": 0.8334276466194346,
      "grad_norm": 0.48469674587249756,
      "learning_rate": 2.8115797943838677e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 29622560,
      "step": 18410
    },
    {
      "epoch": 0.8336539985060776,
      "grad_norm": 0.36315014958381653,
      "learning_rate": 2.810605677321298e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 29630304,
      "step": 18415
    },
    {
      "epoch": 0.8338803503927206,
      "grad_norm": 0.5433885455131531,
      "learning_rate": 2.809631512359428e-05,
      "loss": 0.0877,
      "num_input_tokens_seen": 29638912,
      "step": 18420
    },
    {
      "epoch": 0.8341067022793635,
      "grad_norm": 0.9007452130317688,
      "learning_rate": 2.8086572996484884e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 29646752,
      "step": 18425
    },
    {
      "epoch": 0.8343330541660064,
      "grad_norm": 0.409178763628006,
      "learning_rate": 2.8076830393387143e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 29654880,
      "step": 18430
    },
    {
      "epoch": 0.8345594060526494,
      "grad_norm": 0.5462016463279724,
      "learning_rate": 2.8067087315803497e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 29663200,
      "step": 18435
    },
    {
      "epoch": 0.8347857579392924,
      "grad_norm": 0.5819914937019348,
      "learning_rate": 2.8057343765236433e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 29671040,
      "step": 18440
    },
    {
      "epoch": 0.8350121098259354,
      "grad_norm": 0.4103030860424042,
      "learning_rate": 2.804759974318854e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 29678816,
      "step": 18445
    },
    {
      "epoch": 0.8352384617125784,
      "grad_norm": 0.5898051261901855,
      "learning_rate": 2.8037855251162482e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 29687200,
      "step": 18450
    },
    {
      "epoch": 0.8354648135992213,
      "grad_norm": 0.516173779964447,
      "learning_rate": 2.802811029066096e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 29695424,
      "step": 18455
    },
    {
      "epoch": 0.8356911654858643,
      "grad_norm": 0.30384624004364014,
      "learning_rate": 2.8018364863186764e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 29703232,
      "step": 18460
    },
    {
      "epoch": 0.8359175173725073,
      "grad_norm": 0.5750613212585449,
      "learning_rate": 2.800861897024279e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 29711392,
      "step": 18465
    },
    {
      "epoch": 0.8361438692591503,
      "grad_norm": 0.7365205883979797,
      "learning_rate": 2.799887261333196e-05,
      "loss": 0.1438,
      "num_input_tokens_seen": 29719392,
      "step": 18470
    },
    {
      "epoch": 0.8363702211457933,
      "grad_norm": 1.8716473579406738,
      "learning_rate": 2.798912579395728e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 29726944,
      "step": 18475
    },
    {
      "epoch": 0.8365965730324362,
      "grad_norm": 0.5692981481552124,
      "learning_rate": 2.797937851362185e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 29734688,
      "step": 18480
    },
    {
      "epoch": 0.8368229249190792,
      "grad_norm": 0.48984336853027344,
      "learning_rate": 2.7969630773828802e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 29743008,
      "step": 18485
    },
    {
      "epoch": 0.8370492768057222,
      "grad_norm": 0.5891208052635193,
      "learning_rate": 2.7959882576081382e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 29751424,
      "step": 18490
    },
    {
      "epoch": 0.8372756286923652,
      "grad_norm": 0.500480055809021,
      "learning_rate": 2.795013392188286e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 29759168,
      "step": 18495
    },
    {
      "epoch": 0.8375019805790082,
      "grad_norm": 1.2914009094238281,
      "learning_rate": 2.7940384812736614e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 29767456,
      "step": 18500
    },
    {
      "epoch": 0.8377283324656511,
      "grad_norm": 0.3365727961063385,
      "learning_rate": 2.7930635250146087e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 29776096,
      "step": 18505
    },
    {
      "epoch": 0.837954684352294,
      "grad_norm": 0.6636129021644592,
      "learning_rate": 2.792088523561477e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 29784096,
      "step": 18510
    },
    {
      "epoch": 0.838181036238937,
      "grad_norm": 0.22224964201450348,
      "learning_rate": 2.7911134770646246e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 29791808,
      "step": 18515
    },
    {
      "epoch": 0.83840738812558,
      "grad_norm": 1.1905865669250488,
      "learning_rate": 2.7901383856744157e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 29799424,
      "step": 18520
    },
    {
      "epoch": 0.838633740012223,
      "grad_norm": 1.519930362701416,
      "learning_rate": 2.7891632495412217e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 29807648,
      "step": 18525
    },
    {
      "epoch": 0.838860091898866,
      "grad_norm": 0.47057417035102844,
      "learning_rate": 2.7881880688154205e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 29815232,
      "step": 18530
    },
    {
      "epoch": 0.839086443785509,
      "grad_norm": 0.2802347242832184,
      "learning_rate": 2.7872128436473977e-05,
      "loss": 0.1615,
      "num_input_tokens_seen": 29822848,
      "step": 18535
    },
    {
      "epoch": 0.8393127956721519,
      "grad_norm": 0.33546268939971924,
      "learning_rate": 2.7862375741875448e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 29830240,
      "step": 18540
    },
    {
      "epoch": 0.8395391475587949,
      "grad_norm": 0.668524444103241,
      "learning_rate": 2.785262260586261e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 29838432,
      "step": 18545
    },
    {
      "epoch": 0.8397654994454379,
      "grad_norm": 1.0819449424743652,
      "learning_rate": 2.7842869029939517e-05,
      "loss": 0.1744,
      "num_input_tokens_seen": 29846272,
      "step": 18550
    },
    {
      "epoch": 0.8399918513320809,
      "grad_norm": 0.9236744046211243,
      "learning_rate": 2.7833115015610296e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 29854048,
      "step": 18555
    },
    {
      "epoch": 0.8402182032187239,
      "grad_norm": 0.28525397181510925,
      "learning_rate": 2.7823360564379136e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 29861536,
      "step": 18560
    },
    {
      "epoch": 0.8404445551053668,
      "grad_norm": 0.9303132891654968,
      "learning_rate": 2.7813605677750297e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 29869408,
      "step": 18565
    },
    {
      "epoch": 0.8406709069920097,
      "grad_norm": 0.8261152505874634,
      "learning_rate": 2.7803850357228102e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 29877184,
      "step": 18570
    },
    {
      "epoch": 0.8408972588786527,
      "grad_norm": 0.6676074862480164,
      "learning_rate": 2.779409460431695e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 29885248,
      "step": 18575
    },
    {
      "epoch": 0.8411236107652957,
      "grad_norm": 0.4056226313114166,
      "learning_rate": 2.778433842052129e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 29893504,
      "step": 18580
    },
    {
      "epoch": 0.8413499626519387,
      "grad_norm": 0.7135266661643982,
      "learning_rate": 2.7774581807345664e-05,
      "loss": 0.1158,
      "num_input_tokens_seen": 29901952,
      "step": 18585
    },
    {
      "epoch": 0.8415763145385817,
      "grad_norm": 0.5273367762565613,
      "learning_rate": 2.776482476629465e-05,
      "loss": 0.1366,
      "num_input_tokens_seen": 29909664,
      "step": 18590
    },
    {
      "epoch": 0.8418026664252246,
      "grad_norm": 0.5513474941253662,
      "learning_rate": 2.7755067298872924e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 29917408,
      "step": 18595
    },
    {
      "epoch": 0.8420290183118676,
      "grad_norm": 0.7228512167930603,
      "learning_rate": 2.774530940658518e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 29925408,
      "step": 18600
    },
    {
      "epoch": 0.8420290183118676,
      "eval_loss": 0.1511058658361435,
      "eval_runtime": 404.7603,
      "eval_samples_per_second": 97.023,
      "eval_steps_per_second": 24.256,
      "num_input_tokens_seen": 29925408,
      "step": 18600
    },
    {
      "epoch": 0.8422553701985106,
      "grad_norm": 0.7499992251396179,
      "learning_rate": 2.7735551090936236e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 29933248,
      "step": 18605
    },
    {
      "epoch": 0.8424817220851536,
      "grad_norm": 1.3047114610671997,
      "learning_rate": 2.7725792353430934e-05,
      "loss": 0.12,
      "num_input_tokens_seen": 29941248,
      "step": 18610
    },
    {
      "epoch": 0.8427080739717966,
      "grad_norm": 0.6268692016601562,
      "learning_rate": 2.77160331955742e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 29949728,
      "step": 18615
    },
    {
      "epoch": 0.8429344258584396,
      "grad_norm": 1.0744926929473877,
      "learning_rate": 2.7706273618871008e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 29957440,
      "step": 18620
    },
    {
      "epoch": 0.8431607777450825,
      "grad_norm": 0.6473244428634644,
      "learning_rate": 2.769651362482642e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 29964992,
      "step": 18625
    },
    {
      "epoch": 0.8433871296317255,
      "grad_norm": 0.17818906903266907,
      "learning_rate": 2.768675321494555e-05,
      "loss": 0.1069,
      "num_input_tokens_seen": 29973216,
      "step": 18630
    },
    {
      "epoch": 0.8436134815183685,
      "grad_norm": 0.39935728907585144,
      "learning_rate": 2.7676992390733565e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 29981760,
      "step": 18635
    },
    {
      "epoch": 0.8438398334050115,
      "grad_norm": 1.0822844505310059,
      "learning_rate": 2.766723115369571e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 29989600,
      "step": 18640
    },
    {
      "epoch": 0.8440661852916544,
      "grad_norm": 0.3141982853412628,
      "learning_rate": 2.765746950533729e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 29998016,
      "step": 18645
    },
    {
      "epoch": 0.8442925371782973,
      "grad_norm": 0.7752085328102112,
      "learning_rate": 2.7647707447163684e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 30006048,
      "step": 18650
    },
    {
      "epoch": 0.8445188890649403,
      "grad_norm": 0.17436771094799042,
      "learning_rate": 2.7637944980680315e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 30013824,
      "step": 18655
    },
    {
      "epoch": 0.8447452409515833,
      "grad_norm": 0.6379837393760681,
      "learning_rate": 2.762818210739268e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 30021632,
      "step": 18660
    },
    {
      "epoch": 0.8449715928382263,
      "grad_norm": 0.30444610118865967,
      "learning_rate": 2.7618418828806332e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 30029728,
      "step": 18665
    },
    {
      "epoch": 0.8451979447248693,
      "grad_norm": 0.5260298848152161,
      "learning_rate": 2.76086551464269e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 30037920,
      "step": 18670
    },
    {
      "epoch": 0.8454242966115123,
      "grad_norm": 0.6099506616592407,
      "learning_rate": 2.759889106176006e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 30045632,
      "step": 18675
    },
    {
      "epoch": 0.8456506484981552,
      "grad_norm": 0.5601481795310974,
      "learning_rate": 2.758912657631156e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 30053792,
      "step": 18680
    },
    {
      "epoch": 0.8458770003847982,
      "grad_norm": 0.45437145233154297,
      "learning_rate": 2.7579361691587198e-05,
      "loss": 0.1208,
      "num_input_tokens_seen": 30062240,
      "step": 18685
    },
    {
      "epoch": 0.8461033522714412,
      "grad_norm": 0.8626506924629211,
      "learning_rate": 2.756959640909285e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 30070560,
      "step": 18690
    },
    {
      "epoch": 0.8463297041580842,
      "grad_norm": 0.4933404326438904,
      "learning_rate": 2.7559830730334452e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 30079296,
      "step": 18695
    },
    {
      "epoch": 0.8465560560447272,
      "grad_norm": 0.5279205441474915,
      "learning_rate": 2.7550064656817988e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 30086912,
      "step": 18700
    },
    {
      "epoch": 0.8467824079313702,
      "grad_norm": 0.472445011138916,
      "learning_rate": 2.7540298190049503e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 30094720,
      "step": 18705
    },
    {
      "epoch": 0.847008759818013,
      "grad_norm": 0.5048402547836304,
      "learning_rate": 2.7530531331535107e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 30102848,
      "step": 18710
    },
    {
      "epoch": 0.847235111704656,
      "grad_norm": 0.7250509262084961,
      "learning_rate": 2.752076408278099e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 30110464,
      "step": 18715
    },
    {
      "epoch": 0.847461463591299,
      "grad_norm": 1.1014432907104492,
      "learning_rate": 2.751099644529337e-05,
      "loss": 0.1784,
      "num_input_tokens_seen": 30118464,
      "step": 18720
    },
    {
      "epoch": 0.847687815477942,
      "grad_norm": 0.31567680835723877,
      "learning_rate": 2.7501228420578533e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 30126656,
      "step": 18725
    },
    {
      "epoch": 0.847914167364585,
      "grad_norm": 0.7047285437583923,
      "learning_rate": 2.7491460010142857e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 30134752,
      "step": 18730
    },
    {
      "epoch": 0.8481405192512279,
      "grad_norm": 0.6419621109962463,
      "learning_rate": 2.7481691215492727e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 30142464,
      "step": 18735
    },
    {
      "epoch": 0.8483668711378709,
      "grad_norm": 0.47725793719291687,
      "learning_rate": 2.747192203813463e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 30150144,
      "step": 18740
    },
    {
      "epoch": 0.8485932230245139,
      "grad_norm": 0.7555739879608154,
      "learning_rate": 2.7462152479575087e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 30157984,
      "step": 18745
    },
    {
      "epoch": 0.8488195749111569,
      "grad_norm": 0.5878059267997742,
      "learning_rate": 2.7452382541320697e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 30166496,
      "step": 18750
    },
    {
      "epoch": 0.8490459267977999,
      "grad_norm": 0.7771586179733276,
      "learning_rate": 2.7442612224878096e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 30174432,
      "step": 18755
    },
    {
      "epoch": 0.8492722786844429,
      "grad_norm": 0.6391439437866211,
      "learning_rate": 2.7432841531753994e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 30182592,
      "step": 18760
    },
    {
      "epoch": 0.8494986305710858,
      "grad_norm": 1.076219081878662,
      "learning_rate": 2.7423070463455147e-05,
      "loss": 0.1159,
      "num_input_tokens_seen": 30190240,
      "step": 18765
    },
    {
      "epoch": 0.8497249824577288,
      "grad_norm": 1.1477185487747192,
      "learning_rate": 2.7413299021488397e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 30198048,
      "step": 18770
    },
    {
      "epoch": 0.8499513343443718,
      "grad_norm": 0.5914096236228943,
      "learning_rate": 2.7403527207360615e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 30206112,
      "step": 18775
    },
    {
      "epoch": 0.8501776862310148,
      "grad_norm": 0.8854448199272156,
      "learning_rate": 2.7393755022578722e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 30214400,
      "step": 18780
    },
    {
      "epoch": 0.8504040381176577,
      "grad_norm": 0.8384217023849487,
      "learning_rate": 2.7383982468649714e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 30222560,
      "step": 18785
    },
    {
      "epoch": 0.8506303900043007,
      "grad_norm": 0.6455838680267334,
      "learning_rate": 2.7374209547080665e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 30230560,
      "step": 18790
    },
    {
      "epoch": 0.8508567418909436,
      "grad_norm": 0.3574391007423401,
      "learning_rate": 2.7364436259378663e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 30239040,
      "step": 18795
    },
    {
      "epoch": 0.8510830937775866,
      "grad_norm": 0.5347815752029419,
      "learning_rate": 2.735466260705088e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 30247104,
      "step": 18800
    },
    {
      "epoch": 0.8510830937775866,
      "eval_loss": 0.14857123792171478,
      "eval_runtime": 404.6328,
      "eval_samples_per_second": 97.053,
      "eval_steps_per_second": 24.264,
      "num_input_tokens_seen": 30247104,
      "step": 18800
    },
    {
      "epoch": 0.8513094456642296,
      "grad_norm": 0.42247283458709717,
      "learning_rate": 2.7344888591604524e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 30255136,
      "step": 18805
    },
    {
      "epoch": 0.8515357975508726,
      "grad_norm": 0.4352304935455322,
      "learning_rate": 2.7335114214546893e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 30263072,
      "step": 18810
    },
    {
      "epoch": 0.8517621494375156,
      "grad_norm": 0.6285900473594666,
      "learning_rate": 2.7325339477385293e-05,
      "loss": 0.2229,
      "num_input_tokens_seen": 30271040,
      "step": 18815
    },
    {
      "epoch": 0.8519885013241585,
      "grad_norm": 0.9227632880210876,
      "learning_rate": 2.7315564381627128e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 30279072,
      "step": 18820
    },
    {
      "epoch": 0.8522148532108015,
      "grad_norm": 0.8412203788757324,
      "learning_rate": 2.7305788928779835e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 30287104,
      "step": 18825
    },
    {
      "epoch": 0.8524412050974445,
      "grad_norm": 0.2855248749256134,
      "learning_rate": 2.729601312035091e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 30294912,
      "step": 18830
    },
    {
      "epoch": 0.8526675569840875,
      "grad_norm": 0.8213362097740173,
      "learning_rate": 2.7286236957847915e-05,
      "loss": 0.1728,
      "num_input_tokens_seen": 30304096,
      "step": 18835
    },
    {
      "epoch": 0.8528939088707305,
      "grad_norm": 0.3698965609073639,
      "learning_rate": 2.7276460442778446e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 30312032,
      "step": 18840
    },
    {
      "epoch": 0.8531202607573735,
      "grad_norm": 0.9046363830566406,
      "learning_rate": 2.726668357665017e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 30319904,
      "step": 18845
    },
    {
      "epoch": 0.8533466126440163,
      "grad_norm": 0.7195143699645996,
      "learning_rate": 2.7256906360970808e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 30328000,
      "step": 18850
    },
    {
      "epoch": 0.8535729645306593,
      "grad_norm": 0.41526809334754944,
      "learning_rate": 2.7247128797248117e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 30335776,
      "step": 18855
    },
    {
      "epoch": 0.8537993164173023,
      "grad_norm": 0.5930100083351135,
      "learning_rate": 2.7237350886989925e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 30343680,
      "step": 18860
    },
    {
      "epoch": 0.8540256683039453,
      "grad_norm": 0.5939204096794128,
      "learning_rate": 2.7227572631704107e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 30351680,
      "step": 18865
    },
    {
      "epoch": 0.8542520201905883,
      "grad_norm": 0.6396676301956177,
      "learning_rate": 2.7217794032898596e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 30359488,
      "step": 18870
    },
    {
      "epoch": 0.8544783720772313,
      "grad_norm": 0.3864459693431854,
      "learning_rate": 2.7208015092081384e-05,
      "loss": 0.0948,
      "num_input_tokens_seen": 30367712,
      "step": 18875
    },
    {
      "epoch": 0.8547047239638742,
      "grad_norm": 0.7372753620147705,
      "learning_rate": 2.719823581076049e-05,
      "loss": 0.1913,
      "num_input_tokens_seen": 30375296,
      "step": 18880
    },
    {
      "epoch": 0.8549310758505172,
      "grad_norm": 1.1556782722473145,
      "learning_rate": 2.718845619044401e-05,
      "loss": 0.1748,
      "num_input_tokens_seen": 30382976,
      "step": 18885
    },
    {
      "epoch": 0.8551574277371602,
      "grad_norm": 0.5202690958976746,
      "learning_rate": 2.7178676232640088e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 30390592,
      "step": 18890
    },
    {
      "epoch": 0.8553837796238032,
      "grad_norm": 0.5620453953742981,
      "learning_rate": 2.716889593885691e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 30398976,
      "step": 18895
    },
    {
      "epoch": 0.8556101315104462,
      "grad_norm": 0.9197278022766113,
      "learning_rate": 2.7159115310602716e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 30407200,
      "step": 18900
    },
    {
      "epoch": 0.8558364833970891,
      "grad_norm": 0.3918404281139374,
      "learning_rate": 2.7149334349385814e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 30414848,
      "step": 18905
    },
    {
      "epoch": 0.8560628352837321,
      "grad_norm": 0.6330954432487488,
      "learning_rate": 2.713955305671454e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 30423040,
      "step": 18910
    },
    {
      "epoch": 0.856289187170375,
      "grad_norm": 0.3604823648929596,
      "learning_rate": 2.71297714340973e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 30430624,
      "step": 18915
    },
    {
      "epoch": 0.856515539057018,
      "grad_norm": 0.6842350363731384,
      "learning_rate": 2.7119989483042545e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 30438688,
      "step": 18920
    },
    {
      "epoch": 0.856741890943661,
      "grad_norm": 0.40988802909851074,
      "learning_rate": 2.7110207205058768e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 30446976,
      "step": 18925
    },
    {
      "epoch": 0.856968242830304,
      "grad_norm": 0.5740342140197754,
      "learning_rate": 2.7100424601654517e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 30454432,
      "step": 18930
    },
    {
      "epoch": 0.8571945947169469,
      "grad_norm": 1.0504047870635986,
      "learning_rate": 2.7090641674338403e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 30462752,
      "step": 18935
    },
    {
      "epoch": 0.8574209466035899,
      "grad_norm": 1.459107518196106,
      "learning_rate": 2.7080858424619072e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 30471104,
      "step": 18940
    },
    {
      "epoch": 0.8576472984902329,
      "grad_norm": 1.1803843975067139,
      "learning_rate": 2.707107485400521e-05,
      "loss": 0.18,
      "num_input_tokens_seen": 30478656,
      "step": 18945
    },
    {
      "epoch": 0.8578736503768759,
      "grad_norm": 0.357650488615036,
      "learning_rate": 2.7061290964005586e-05,
      "loss": 0.116,
      "num_input_tokens_seen": 30487104,
      "step": 18950
    },
    {
      "epoch": 0.8581000022635189,
      "grad_norm": 0.5782884955406189,
      "learning_rate": 2.7051506756129e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 30494624,
      "step": 18955
    },
    {
      "epoch": 0.8583263541501619,
      "grad_norm": 0.4193708300590515,
      "learning_rate": 2.704172223188428e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 30503072,
      "step": 18960
    },
    {
      "epoch": 0.8585527060368048,
      "grad_norm": 0.5943185687065125,
      "learning_rate": 2.7031937392780334e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 30511008,
      "step": 18965
    },
    {
      "epoch": 0.8587790579234478,
      "grad_norm": 0.5101448893547058,
      "learning_rate": 2.702215224032611e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 30518752,
      "step": 18970
    },
    {
      "epoch": 0.8590054098100908,
      "grad_norm": 0.6968523859977722,
      "learning_rate": 2.70123667760306e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 30526560,
      "step": 18975
    },
    {
      "epoch": 0.8592317616967338,
      "grad_norm": 0.3226737380027771,
      "learning_rate": 2.7002581001402845e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 30533952,
      "step": 18980
    },
    {
      "epoch": 0.8594581135833768,
      "grad_norm": 0.6673051714897156,
      "learning_rate": 2.6992794917951923e-05,
      "loss": 0.1658,
      "num_input_tokens_seen": 30541792,
      "step": 18985
    },
    {
      "epoch": 0.8596844654700196,
      "grad_norm": 1.52040696144104,
      "learning_rate": 2.6983008527187e-05,
      "loss": 0.1851,
      "num_input_tokens_seen": 30550464,
      "step": 18990
    },
    {
      "epoch": 0.8599108173566626,
      "grad_norm": 0.7192667126655579,
      "learning_rate": 2.697322183061723e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 30558432,
      "step": 18995
    },
    {
      "epoch": 0.8601371692433056,
      "grad_norm": 0.7381038069725037,
      "learning_rate": 2.696343482975186e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 30566528,
      "step": 19000
    },
    {
      "epoch": 0.8601371692433056,
      "eval_loss": 0.14825788140296936,
      "eval_runtime": 405.165,
      "eval_samples_per_second": 96.926,
      "eval_steps_per_second": 24.232,
      "num_input_tokens_seen": 30566528,
      "step": 19000
    },
    {
      "epoch": 0.8603635211299486,
      "grad_norm": 0.6829057931900024,
      "learning_rate": 2.695364752610016e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 30574848,
      "step": 19005
    },
    {
      "epoch": 0.8605898730165916,
      "grad_norm": 0.7672156095504761,
      "learning_rate": 2.6943859921171467e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 30582880,
      "step": 19010
    },
    {
      "epoch": 0.8608162249032346,
      "grad_norm": 0.7964276671409607,
      "learning_rate": 2.6934072016475143e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 30590304,
      "step": 19015
    },
    {
      "epoch": 0.8610425767898775,
      "grad_norm": 0.7468669414520264,
      "learning_rate": 2.6924283813520606e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 30598912,
      "step": 19020
    },
    {
      "epoch": 0.8612689286765205,
      "grad_norm": 0.6427022218704224,
      "learning_rate": 2.691449531381733e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 30606816,
      "step": 19025
    },
    {
      "epoch": 0.8614952805631635,
      "grad_norm": 0.6696485280990601,
      "learning_rate": 2.6904706518874816e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 30614944,
      "step": 19030
    },
    {
      "epoch": 0.8617216324498065,
      "grad_norm": 1.0141963958740234,
      "learning_rate": 2.6894917430202615e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 30623488,
      "step": 19035
    },
    {
      "epoch": 0.8619479843364495,
      "grad_norm": 0.5719294548034668,
      "learning_rate": 2.6885128049310343e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 30631872,
      "step": 19040
    },
    {
      "epoch": 0.8621743362230925,
      "grad_norm": 1.0180788040161133,
      "learning_rate": 2.687533837770762e-05,
      "loss": 0.1959,
      "num_input_tokens_seen": 30640320,
      "step": 19045
    },
    {
      "epoch": 0.8624006881097354,
      "grad_norm": 0.7529134750366211,
      "learning_rate": 2.6865548416904162e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 30648768,
      "step": 19050
    },
    {
      "epoch": 0.8626270399963784,
      "grad_norm": 0.4654097557067871,
      "learning_rate": 2.68557581684097e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 30656544,
      "step": 19055
    },
    {
      "epoch": 0.8628533918830213,
      "grad_norm": 0.50648432970047,
      "learning_rate": 2.6845967633733998e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 30664704,
      "step": 19060
    },
    {
      "epoch": 0.8630797437696643,
      "grad_norm": 0.7495653629302979,
      "learning_rate": 2.683617681438689e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 30672640,
      "step": 19065
    },
    {
      "epoch": 0.8633060956563073,
      "grad_norm": 0.49101418256759644,
      "learning_rate": 2.682638571187825e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 30680256,
      "step": 19070
    },
    {
      "epoch": 0.8635324475429502,
      "grad_norm": 0.6569118499755859,
      "learning_rate": 2.6816594327717976e-05,
      "loss": 0.2212,
      "num_input_tokens_seen": 30688032,
      "step": 19075
    },
    {
      "epoch": 0.8637587994295932,
      "grad_norm": 0.6588127613067627,
      "learning_rate": 2.680680266341603e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 30695808,
      "step": 19080
    },
    {
      "epoch": 0.8639851513162362,
      "grad_norm": 0.8050307631492615,
      "learning_rate": 2.67970107204824e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 30703680,
      "step": 19085
    },
    {
      "epoch": 0.8642115032028792,
      "grad_norm": 1.135724663734436,
      "learning_rate": 2.6787218500427142e-05,
      "loss": 0.1684,
      "num_input_tokens_seen": 30712544,
      "step": 19090
    },
    {
      "epoch": 0.8644378550895222,
      "grad_norm": 0.7457836270332336,
      "learning_rate": 2.6777426004760332e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 30720064,
      "step": 19095
    },
    {
      "epoch": 0.8646642069761652,
      "grad_norm": 1.227737307548523,
      "learning_rate": 2.6767633234992094e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 30727296,
      "step": 19100
    },
    {
      "epoch": 0.8648905588628081,
      "grad_norm": 0.404965341091156,
      "learning_rate": 2.6757840192632598e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 30734752,
      "step": 19105
    },
    {
      "epoch": 0.8651169107494511,
      "grad_norm": 0.59553062915802,
      "learning_rate": 2.6748046879192052e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 30742624,
      "step": 19110
    },
    {
      "epoch": 0.8653432626360941,
      "grad_norm": 0.8053215742111206,
      "learning_rate": 2.673825329618071e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 30750656,
      "step": 19115
    },
    {
      "epoch": 0.8655696145227371,
      "grad_norm": 0.4302161931991577,
      "learning_rate": 2.6728459445108866e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 30759104,
      "step": 19120
    },
    {
      "epoch": 0.8657959664093801,
      "grad_norm": 0.571219265460968,
      "learning_rate": 2.6718665327486854e-05,
      "loss": 0.1069,
      "num_input_tokens_seen": 30767008,
      "step": 19125
    },
    {
      "epoch": 0.866022318296023,
      "grad_norm": 0.6309125423431396,
      "learning_rate": 2.6708870944825048e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 30774944,
      "step": 19130
    },
    {
      "epoch": 0.8662486701826659,
      "grad_norm": 0.967564046382904,
      "learning_rate": 2.6699076298633874e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 30783072,
      "step": 19135
    },
    {
      "epoch": 0.8664750220693089,
      "grad_norm": 0.35560888051986694,
      "learning_rate": 2.6689281390423788e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 30791264,
      "step": 19140
    },
    {
      "epoch": 0.8667013739559519,
      "grad_norm": 0.6104452013969421,
      "learning_rate": 2.667948622170527e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 30799392,
      "step": 19145
    },
    {
      "epoch": 0.8669277258425949,
      "grad_norm": 0.47610175609588623,
      "learning_rate": 2.6669690793988873e-05,
      "loss": 0.1697,
      "num_input_tokens_seen": 30807392,
      "step": 19150
    },
    {
      "epoch": 0.8671540777292379,
      "grad_norm": 0.9513158202171326,
      "learning_rate": 2.665989510878518e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 30815488,
      "step": 19155
    },
    {
      "epoch": 0.8673804296158808,
      "grad_norm": 0.717873752117157,
      "learning_rate": 2.6650099167604793e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 30823392,
      "step": 19160
    },
    {
      "epoch": 0.8676067815025238,
      "grad_norm": 0.5439401865005493,
      "learning_rate": 2.6640302971958376e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 30831264,
      "step": 19165
    },
    {
      "epoch": 0.8678331333891668,
      "grad_norm": 0.3348836302757263,
      "learning_rate": 2.6630506523356635e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 30838720,
      "step": 19170
    },
    {
      "epoch": 0.8680594852758098,
      "grad_norm": 0.5901443958282471,
      "learning_rate": 2.6620709823310297e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 30846432,
      "step": 19175
    },
    {
      "epoch": 0.8682858371624528,
      "grad_norm": 0.3389987647533417,
      "learning_rate": 2.661091287333014e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 30854848,
      "step": 19180
    },
    {
      "epoch": 0.8685121890490958,
      "grad_norm": 0.5234993696212769,
      "learning_rate": 2.660111567492696e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 30862880,
      "step": 19185
    },
    {
      "epoch": 0.8687385409357387,
      "grad_norm": 0.2760854661464691,
      "learning_rate": 2.6591318229611635e-05,
      "loss": 0.0878,
      "num_input_tokens_seen": 30870880,
      "step": 19190
    },
    {
      "epoch": 0.8689648928223817,
      "grad_norm": 0.47301143407821655,
      "learning_rate": 2.6581520538895037e-05,
      "loss": 0.1816,
      "num_input_tokens_seen": 30879744,
      "step": 19195
    },
    {
      "epoch": 0.8691912447090246,
      "grad_norm": 0.6016550660133362,
      "learning_rate": 2.6571722604288102e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 30887744,
      "step": 19200
    },
    {
      "epoch": 0.8691912447090246,
      "eval_loss": 0.14810490608215332,
      "eval_runtime": 404.571,
      "eval_samples_per_second": 97.068,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 30887744,
      "step": 19200
    },
    {
      "epoch": 0.8694175965956676,
      "grad_norm": 0.4317912757396698,
      "learning_rate": 2.656192442730179e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 30896192,
      "step": 19205
    },
    {
      "epoch": 0.8696439484823106,
      "grad_norm": 0.3407076299190521,
      "learning_rate": 2.6552126009447098e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 30904160,
      "step": 19210
    },
    {
      "epoch": 0.8698703003689536,
      "grad_norm": 0.5667308568954468,
      "learning_rate": 2.654232735223507e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 30912128,
      "step": 19215
    },
    {
      "epoch": 0.8700966522555965,
      "grad_norm": 0.45534199476242065,
      "learning_rate": 2.6532528457176787e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 30920000,
      "step": 19220
    },
    {
      "epoch": 0.8703230041422395,
      "grad_norm": 0.8670610189437866,
      "learning_rate": 2.6522729325783348e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 30928352,
      "step": 19225
    },
    {
      "epoch": 0.8705493560288825,
      "grad_norm": 0.6899641752243042,
      "learning_rate": 2.6512929959565914e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 30936896,
      "step": 19230
    },
    {
      "epoch": 0.8707757079155255,
      "grad_norm": 0.41240978240966797,
      "learning_rate": 2.6503130360035673e-05,
      "loss": 0.1948,
      "num_input_tokens_seen": 30944480,
      "step": 19235
    },
    {
      "epoch": 0.8710020598021685,
      "grad_norm": 0.695820689201355,
      "learning_rate": 2.6493330528703835e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 30953120,
      "step": 19240
    },
    {
      "epoch": 0.8712284116888114,
      "grad_norm": 0.3652840554714203,
      "learning_rate": 2.648353046708167e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 30960896,
      "step": 19245
    },
    {
      "epoch": 0.8714547635754544,
      "grad_norm": 0.5245837569236755,
      "learning_rate": 2.647373017668046e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 30969056,
      "step": 19250
    },
    {
      "epoch": 0.8716811154620974,
      "grad_norm": 0.4203015863895416,
      "learning_rate": 2.6463929659011537e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 30976512,
      "step": 19255
    },
    {
      "epoch": 0.8719074673487404,
      "grad_norm": 1.4081171751022339,
      "learning_rate": 2.6454128915586262e-05,
      "loss": 0.1847,
      "num_input_tokens_seen": 30984608,
      "step": 19260
    },
    {
      "epoch": 0.8721338192353834,
      "grad_norm": 0.6147666573524475,
      "learning_rate": 2.6444327947916036e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 30992032,
      "step": 19265
    },
    {
      "epoch": 0.8723601711220264,
      "grad_norm": 0.6403318643569946,
      "learning_rate": 2.6434526757512292e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 30999968,
      "step": 19270
    },
    {
      "epoch": 0.8725865230086692,
      "grad_norm": 0.4392205774784088,
      "learning_rate": 2.6424725345886486e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 31007872,
      "step": 19275
    },
    {
      "epoch": 0.8728128748953122,
      "grad_norm": 0.39733195304870605,
      "learning_rate": 2.641492371455014e-05,
      "loss": 0.0974,
      "num_input_tokens_seen": 31015744,
      "step": 19280
    },
    {
      "epoch": 0.8730392267819552,
      "grad_norm": 0.6604623198509216,
      "learning_rate": 2.640512186501477e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 31023456,
      "step": 19285
    },
    {
      "epoch": 0.8732655786685982,
      "grad_norm": 0.20833130180835724,
      "learning_rate": 2.639531979879195e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 31031712,
      "step": 19290
    },
    {
      "epoch": 0.8734919305552412,
      "grad_norm": 0.5771517157554626,
      "learning_rate": 2.638551751739328e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 31040160,
      "step": 19295
    },
    {
      "epoch": 0.8737182824418841,
      "grad_norm": 0.6055824160575867,
      "learning_rate": 2.6375715022330404e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 31047776,
      "step": 19300
    },
    {
      "epoch": 0.8739446343285271,
      "grad_norm": 0.4899052381515503,
      "learning_rate": 2.6365912315114976e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 31056448,
      "step": 19305
    },
    {
      "epoch": 0.8741709862151701,
      "grad_norm": 0.912980318069458,
      "learning_rate": 2.6356109397258704e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 31064736,
      "step": 19310
    },
    {
      "epoch": 0.8743973381018131,
      "grad_norm": 1.118869662284851,
      "learning_rate": 2.6346306270273325e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 31072160,
      "step": 19315
    },
    {
      "epoch": 0.8746236899884561,
      "grad_norm": 0.3981574475765228,
      "learning_rate": 2.6336502935670608e-05,
      "loss": 0.1869,
      "num_input_tokens_seen": 31080256,
      "step": 19320
    },
    {
      "epoch": 0.8748500418750991,
      "grad_norm": 0.6244338750839233,
      "learning_rate": 2.6326699394962333e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 31088384,
      "step": 19325
    },
    {
      "epoch": 0.875076393761742,
      "grad_norm": 0.8893200159072876,
      "learning_rate": 2.6316895649660334e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 31096576,
      "step": 19330
    },
    {
      "epoch": 0.875302745648385,
      "grad_norm": 0.5165644288063049,
      "learning_rate": 2.6307091701276486e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 31104448,
      "step": 19335
    },
    {
      "epoch": 0.875529097535028,
      "grad_norm": 0.5489224791526794,
      "learning_rate": 2.629728755132267e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 31112288,
      "step": 19340
    },
    {
      "epoch": 0.8757554494216709,
      "grad_norm": 0.6723659634590149,
      "learning_rate": 2.628748320131081e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 31120320,
      "step": 19345
    },
    {
      "epoch": 0.8759818013083139,
      "grad_norm": 0.6007727980613708,
      "learning_rate": 2.6277678652752856e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 31128096,
      "step": 19350
    },
    {
      "epoch": 0.8762081531949569,
      "grad_norm": 0.954450786113739,
      "learning_rate": 2.6267873907160807e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 31136320,
      "step": 19355
    },
    {
      "epoch": 0.8764345050815998,
      "grad_norm": 0.7890543937683105,
      "learning_rate": 2.6258068966046668e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 31144640,
      "step": 19360
    },
    {
      "epoch": 0.8766608569682428,
      "grad_norm": 0.4908427894115448,
      "learning_rate": 2.6248263830922475e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 31152160,
      "step": 19365
    },
    {
      "epoch": 0.8768872088548858,
      "grad_norm": 0.5837008953094482,
      "learning_rate": 2.6238458503300318e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 31160096,
      "step": 19370
    },
    {
      "epoch": 0.8771135607415288,
      "grad_norm": 0.6912423372268677,
      "learning_rate": 2.6228652984692292e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 31168160,
      "step": 19375
    },
    {
      "epoch": 0.8773399126281718,
      "grad_norm": 1.0927319526672363,
      "learning_rate": 2.621884727661054e-05,
      "loss": 0.1761,
      "num_input_tokens_seen": 31175936,
      "step": 19380
    },
    {
      "epoch": 0.8775662645148147,
      "grad_norm": 1.0411303043365479,
      "learning_rate": 2.6209041380567222e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 31183936,
      "step": 19385
    },
    {
      "epoch": 0.8777926164014577,
      "grad_norm": 0.5834040641784668,
      "learning_rate": 2.6199235298074527e-05,
      "loss": 0.2063,
      "num_input_tokens_seen": 31191616,
      "step": 19390
    },
    {
      "epoch": 0.8780189682881007,
      "grad_norm": 0.4402243494987488,
      "learning_rate": 2.618942903064468e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 31199840,
      "step": 19395
    },
    {
      "epoch": 0.8782453201747437,
      "grad_norm": 0.7131270170211792,
      "learning_rate": 2.6179622579789932e-05,
      "loss": 0.0964,
      "num_input_tokens_seen": 31208352,
      "step": 19400
    },
    {
      "epoch": 0.8782453201747437,
      "eval_loss": 0.14837218821048737,
      "eval_runtime": 404.7544,
      "eval_samples_per_second": 97.024,
      "eval_steps_per_second": 24.257,
      "num_input_tokens_seen": 31208352,
      "step": 19400
    },
    {
      "epoch": 0.8784716720613867,
      "grad_norm": 0.23325829207897186,
      "learning_rate": 2.6169815947022553e-05,
      "loss": 0.1079,
      "num_input_tokens_seen": 31216640,
      "step": 19405
    },
    {
      "epoch": 0.8786980239480296,
      "grad_norm": 0.8519164323806763,
      "learning_rate": 2.6160009133854853e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 31224672,
      "step": 19410
    },
    {
      "epoch": 0.8789243758346725,
      "grad_norm": 1.1555287837982178,
      "learning_rate": 2.6150202141799168e-05,
      "loss": 0.1718,
      "num_input_tokens_seen": 31232544,
      "step": 19415
    },
    {
      "epoch": 0.8791507277213155,
      "grad_norm": 0.5413447022438049,
      "learning_rate": 2.614039497236786e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 31240544,
      "step": 19420
    },
    {
      "epoch": 0.8793770796079585,
      "grad_norm": 0.636451244354248,
      "learning_rate": 2.6130587627073315e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 31248448,
      "step": 19425
    },
    {
      "epoch": 0.8796034314946015,
      "grad_norm": 1.1535990238189697,
      "learning_rate": 2.6120780107427956e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 31256704,
      "step": 19430
    },
    {
      "epoch": 0.8798297833812445,
      "grad_norm": 1.355002760887146,
      "learning_rate": 2.6110972414944214e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 31264896,
      "step": 19435
    },
    {
      "epoch": 0.8800561352678875,
      "grad_norm": 2.455343723297119,
      "learning_rate": 2.6101164551134565e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 31272480,
      "step": 19440
    },
    {
      "epoch": 0.8802824871545304,
      "grad_norm": 1.0968698263168335,
      "learning_rate": 2.6091356517511505e-05,
      "loss": 0.1704,
      "num_input_tokens_seen": 31280096,
      "step": 19445
    },
    {
      "epoch": 0.8805088390411734,
      "grad_norm": 0.5890365242958069,
      "learning_rate": 2.608154831558755e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 31288032,
      "step": 19450
    },
    {
      "epoch": 0.8807351909278164,
      "grad_norm": 0.911899745464325,
      "learning_rate": 2.607173994687526e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 31296096,
      "step": 19455
    },
    {
      "epoch": 0.8809615428144594,
      "grad_norm": 1.1601629257202148,
      "learning_rate": 2.6061931412887196e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 31303712,
      "step": 19460
    },
    {
      "epoch": 0.8811878947011024,
      "grad_norm": 0.43302100896835327,
      "learning_rate": 2.6052122715135973e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 31311424,
      "step": 19465
    },
    {
      "epoch": 0.8814142465877453,
      "grad_norm": 0.8062325119972229,
      "learning_rate": 2.60423138551342e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 31319168,
      "step": 19470
    },
    {
      "epoch": 0.8816405984743882,
      "grad_norm": 0.729546844959259,
      "learning_rate": 2.6032504834394527e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 31327360,
      "step": 19475
    },
    {
      "epoch": 0.8818669503610312,
      "grad_norm": 0.529645562171936,
      "learning_rate": 2.602269565442964e-05,
      "loss": 0.2062,
      "num_input_tokens_seen": 31335072,
      "step": 19480
    },
    {
      "epoch": 0.8820933022476742,
      "grad_norm": 1.2075839042663574,
      "learning_rate": 2.6012886316752227e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 31343072,
      "step": 19485
    },
    {
      "epoch": 0.8823196541343172,
      "grad_norm": 1.1100366115570068,
      "learning_rate": 2.6003076822875018e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 31350688,
      "step": 19490
    },
    {
      "epoch": 0.8825460060209602,
      "grad_norm": 0.855728030204773,
      "learning_rate": 2.5993267174310755e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 31358176,
      "step": 19495
    },
    {
      "epoch": 0.8827723579076031,
      "grad_norm": 0.7158449292182922,
      "learning_rate": 2.5983457372572218e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 31366304,
      "step": 19500
    },
    {
      "epoch": 0.8829987097942461,
      "grad_norm": 0.4576990306377411,
      "learning_rate": 2.597364741917219e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 31374336,
      "step": 19505
    },
    {
      "epoch": 0.8832250616808891,
      "grad_norm": 0.5184422135353088,
      "learning_rate": 2.5963837315623492e-05,
      "loss": 0.1578,
      "num_input_tokens_seen": 31382688,
      "step": 19510
    },
    {
      "epoch": 0.8834514135675321,
      "grad_norm": 0.5337323546409607,
      "learning_rate": 2.595402706343897e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 31390592,
      "step": 19515
    },
    {
      "epoch": 0.8836777654541751,
      "grad_norm": 0.30122581124305725,
      "learning_rate": 2.594421666413148e-05,
      "loss": 0.1196,
      "num_input_tokens_seen": 31398848,
      "step": 19520
    },
    {
      "epoch": 0.8839041173408181,
      "grad_norm": 0.5701379179954529,
      "learning_rate": 2.5934406119213928e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 31406848,
      "step": 19525
    },
    {
      "epoch": 0.884130469227461,
      "grad_norm": 0.5718246698379517,
      "learning_rate": 2.5924595430199193e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 31414752,
      "step": 19530
    },
    {
      "epoch": 0.884356821114104,
      "grad_norm": 0.49270832538604736,
      "learning_rate": 2.5914784598600238e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 31422560,
      "step": 19535
    },
    {
      "epoch": 0.884583173000747,
      "grad_norm": 0.9058638215065002,
      "learning_rate": 2.5904973625930002e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 31430624,
      "step": 19540
    },
    {
      "epoch": 0.88480952488739,
      "grad_norm": 0.7294504046440125,
      "learning_rate": 2.5895162513701456e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 31439168,
      "step": 19545
    },
    {
      "epoch": 0.885035876774033,
      "grad_norm": 0.7866982221603394,
      "learning_rate": 2.5885351263427593e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 31447328,
      "step": 19550
    },
    {
      "epoch": 0.8852622286606758,
      "grad_norm": 1.1215622425079346,
      "learning_rate": 2.5875539876621448e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 31455136,
      "step": 19555
    },
    {
      "epoch": 0.8854885805473188,
      "grad_norm": 1.2209864854812622,
      "learning_rate": 2.586572835479605e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 31462848,
      "step": 19560
    },
    {
      "epoch": 0.8857149324339618,
      "grad_norm": 0.9437758922576904,
      "learning_rate": 2.585591669946446e-05,
      "loss": 0.175,
      "num_input_tokens_seen": 31470944,
      "step": 19565
    },
    {
      "epoch": 0.8859412843206048,
      "grad_norm": 0.7108607888221741,
      "learning_rate": 2.5846104912139756e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 31479040,
      "step": 19570
    },
    {
      "epoch": 0.8861676362072478,
      "grad_norm": 0.4027706980705261,
      "learning_rate": 2.583629299433505e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 31487200,
      "step": 19575
    },
    {
      "epoch": 0.8863939880938908,
      "grad_norm": 0.9842677712440491,
      "learning_rate": 2.582648094756345e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 31494752,
      "step": 19580
    },
    {
      "epoch": 0.8866203399805337,
      "grad_norm": 0.40166959166526794,
      "learning_rate": 2.5816668773338098e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 31502752,
      "step": 19585
    },
    {
      "epoch": 0.8868466918671767,
      "grad_norm": 1.0087871551513672,
      "learning_rate": 2.580685647317216e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 31511136,
      "step": 19590
    },
    {
      "epoch": 0.8870730437538197,
      "grad_norm": 0.608591616153717,
      "learning_rate": 2.5797044048578818e-05,
      "loss": 0.1259,
      "num_input_tokens_seen": 31518912,
      "step": 19595
    },
    {
      "epoch": 0.8872993956404627,
      "grad_norm": 0.4335019588470459,
      "learning_rate": 2.5787231501071262e-05,
      "loss": 0.1624,
      "num_input_tokens_seen": 31526720,
      "step": 19600
    },
    {
      "epoch": 0.8872993956404627,
      "eval_loss": 0.1481919288635254,
      "eval_runtime": 404.6377,
      "eval_samples_per_second": 97.052,
      "eval_steps_per_second": 24.264,
      "num_input_tokens_seen": 31526720,
      "step": 19600
    },
    {
      "epoch": 0.8875257475271057,
      "grad_norm": 0.4828341007232666,
      "learning_rate": 2.577741883216272e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 31534720,
      "step": 19605
    },
    {
      "epoch": 0.8877520994137487,
      "grad_norm": 1.3256455659866333,
      "learning_rate": 2.576760604336642e-05,
      "loss": 0.2014,
      "num_input_tokens_seen": 31542656,
      "step": 19610
    },
    {
      "epoch": 0.8879784513003915,
      "grad_norm": 0.41895994544029236,
      "learning_rate": 2.575779313619563e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 31550272,
      "step": 19615
    },
    {
      "epoch": 0.8882048031870345,
      "grad_norm": 0.2992998957633972,
      "learning_rate": 2.5747980112163605e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 31558368,
      "step": 19620
    },
    {
      "epoch": 0.8884311550736775,
      "grad_norm": 0.7037586569786072,
      "learning_rate": 2.5738166972783656e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 31566016,
      "step": 19625
    },
    {
      "epoch": 0.8886575069603205,
      "grad_norm": 1.2402247190475464,
      "learning_rate": 2.5728353719569075e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 31573856,
      "step": 19630
    },
    {
      "epoch": 0.8888838588469635,
      "grad_norm": 0.47784486413002014,
      "learning_rate": 2.57185403540332e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 31582048,
      "step": 19635
    },
    {
      "epoch": 0.8891102107336064,
      "grad_norm": 0.5345825552940369,
      "learning_rate": 2.5708726877689375e-05,
      "loss": 0.1806,
      "num_input_tokens_seen": 31590368,
      "step": 19640
    },
    {
      "epoch": 0.8893365626202494,
      "grad_norm": 0.6149384379386902,
      "learning_rate": 2.5698913292050964e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 31598528,
      "step": 19645
    },
    {
      "epoch": 0.8895629145068924,
      "grad_norm": 1.1372239589691162,
      "learning_rate": 2.568909959863133e-05,
      "loss": 0.1988,
      "num_input_tokens_seen": 31606368,
      "step": 19650
    },
    {
      "epoch": 0.8897892663935354,
      "grad_norm": 0.424769788980484,
      "learning_rate": 2.5679285798943887e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 31614240,
      "step": 19655
    },
    {
      "epoch": 0.8900156182801784,
      "grad_norm": 0.7860288023948669,
      "learning_rate": 2.5669471894502035e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 31621824,
      "step": 19660
    },
    {
      "epoch": 0.8902419701668214,
      "grad_norm": 0.5314584970474243,
      "learning_rate": 2.56596578868192e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 31630144,
      "step": 19665
    },
    {
      "epoch": 0.8904683220534643,
      "grad_norm": 0.9206334352493286,
      "learning_rate": 2.564984377740883e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 31638848,
      "step": 19670
    },
    {
      "epoch": 0.8906946739401073,
      "grad_norm": 0.623947024345398,
      "learning_rate": 2.564002956778438e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 31646848,
      "step": 19675
    },
    {
      "epoch": 0.8909210258267503,
      "grad_norm": 0.46665212512016296,
      "learning_rate": 2.563021525945934e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 31654656,
      "step": 19680
    },
    {
      "epoch": 0.8911473777133933,
      "grad_norm": 0.5559872984886169,
      "learning_rate": 2.562040085394718e-05,
      "loss": 0.154,
      "num_input_tokens_seen": 31662656,
      "step": 19685
    },
    {
      "epoch": 0.8913737296000362,
      "grad_norm": 0.8272870182991028,
      "learning_rate": 2.56105863527614e-05,
      "loss": 0.205,
      "num_input_tokens_seen": 31670656,
      "step": 19690
    },
    {
      "epoch": 0.8916000814866792,
      "grad_norm": 1.925271987915039,
      "learning_rate": 2.5600771757415548e-05,
      "loss": 0.1758,
      "num_input_tokens_seen": 31679200,
      "step": 19695
    },
    {
      "epoch": 0.8918264333733221,
      "grad_norm": 0.7341111898422241,
      "learning_rate": 2.5590957069423134e-05,
      "loss": 0.1835,
      "num_input_tokens_seen": 31686688,
      "step": 19700
    },
    {
      "epoch": 0.8920527852599651,
      "grad_norm": 0.32960084080696106,
      "learning_rate": 2.5581142290297716e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 31694752,
      "step": 19705
    },
    {
      "epoch": 0.8922791371466081,
      "grad_norm": 0.38590991497039795,
      "learning_rate": 2.557132742155285e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 31703008,
      "step": 19710
    },
    {
      "epoch": 0.8925054890332511,
      "grad_norm": 1.295597791671753,
      "learning_rate": 2.556151246470212e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 31710528,
      "step": 19715
    },
    {
      "epoch": 0.8927318409198941,
      "grad_norm": 1.0001521110534668,
      "learning_rate": 2.5551697421259114e-05,
      "loss": 0.2175,
      "num_input_tokens_seen": 31718464,
      "step": 19720
    },
    {
      "epoch": 0.892958192806537,
      "grad_norm": 0.4628969430923462,
      "learning_rate": 2.554188229273743e-05,
      "loss": 0.1195,
      "num_input_tokens_seen": 31727072,
      "step": 19725
    },
    {
      "epoch": 0.89318454469318,
      "grad_norm": 0.38290101289749146,
      "learning_rate": 2.5532067080650678e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 31735264,
      "step": 19730
    },
    {
      "epoch": 0.893410896579823,
      "grad_norm": 0.9497001767158508,
      "learning_rate": 2.55222517865125e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 31743232,
      "step": 19735
    },
    {
      "epoch": 0.893637248466466,
      "grad_norm": 0.4096758961677551,
      "learning_rate": 2.5512436411836538e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 31751392,
      "step": 19740
    },
    {
      "epoch": 0.893863600353109,
      "grad_norm": 1.5473111867904663,
      "learning_rate": 2.5502620958136443e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 31759712,
      "step": 19745
    },
    {
      "epoch": 0.894089952239752,
      "grad_norm": 0.9013803601264954,
      "learning_rate": 2.5492805426925874e-05,
      "loss": 0.1442,
      "num_input_tokens_seen": 31767424,
      "step": 19750
    },
    {
      "epoch": 0.8943163041263948,
      "grad_norm": 0.5816416144371033,
      "learning_rate": 2.5482989819718523e-05,
      "loss": 0.1144,
      "num_input_tokens_seen": 31775136,
      "step": 19755
    },
    {
      "epoch": 0.8945426560130378,
      "grad_norm": 0.9253882765769958,
      "learning_rate": 2.5473174138028065e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 31783264,
      "step": 19760
    },
    {
      "epoch": 0.8947690078996808,
      "grad_norm": 0.9880849123001099,
      "learning_rate": 2.5463358383368212e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 31791008,
      "step": 19765
    },
    {
      "epoch": 0.8949953597863238,
      "grad_norm": 0.3891991078853607,
      "learning_rate": 2.545354255725267e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 31799360,
      "step": 19770
    },
    {
      "epoch": 0.8952217116729668,
      "grad_norm": 0.7241577506065369,
      "learning_rate": 2.5443726661195165e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 31807936,
      "step": 19775
    },
    {
      "epoch": 0.8954480635596098,
      "grad_norm": 0.4833892285823822,
      "learning_rate": 2.543391069670944e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 31816160,
      "step": 19780
    },
    {
      "epoch": 0.8956744154462527,
      "grad_norm": 0.3556186556816101,
      "learning_rate": 2.5424094665309228e-05,
      "loss": 0.1686,
      "num_input_tokens_seen": 31824416,
      "step": 19785
    },
    {
      "epoch": 0.8959007673328957,
      "grad_norm": 0.36628326773643494,
      "learning_rate": 2.5414278568508292e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 31832640,
      "step": 19790
    },
    {
      "epoch": 0.8961271192195387,
      "grad_norm": 0.4539414942264557,
      "learning_rate": 2.540446240782039e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 31840672,
      "step": 19795
    },
    {
      "epoch": 0.8963534711061817,
      "grad_norm": 1.3659265041351318,
      "learning_rate": 2.5394646184759307e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 31848736,
      "step": 19800
    },
    {
      "epoch": 0.8963534711061817,
      "eval_loss": 0.14741070568561554,
      "eval_runtime": 404.5736,
      "eval_samples_per_second": 97.068,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 31848736,
      "step": 19800
    },
    {
      "epoch": 0.8965798229928247,
      "grad_norm": 0.6910096406936646,
      "learning_rate": 2.538482990083882e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 31857440,
      "step": 19805
    },
    {
      "epoch": 0.8968061748794676,
      "grad_norm": 0.7758543491363525,
      "learning_rate": 2.5375013557572725e-05,
      "loss": 0.1727,
      "num_input_tokens_seen": 31866048,
      "step": 19810
    },
    {
      "epoch": 0.8970325267661106,
      "grad_norm": 0.5629045367240906,
      "learning_rate": 2.536519715647483e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 31873632,
      "step": 19815
    },
    {
      "epoch": 0.8972588786527536,
      "grad_norm": 0.5835235714912415,
      "learning_rate": 2.535538069905894e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 31882432,
      "step": 19820
    },
    {
      "epoch": 0.8974852305393966,
      "grad_norm": 0.48053404688835144,
      "learning_rate": 2.534556418683888e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 31890944,
      "step": 19825
    },
    {
      "epoch": 0.8977115824260395,
      "grad_norm": 0.7207124829292297,
      "learning_rate": 2.5335747621328486e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 31898784,
      "step": 19830
    },
    {
      "epoch": 0.8979379343126825,
      "grad_norm": 0.9784935116767883,
      "learning_rate": 2.5325931004041586e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 31906912,
      "step": 19835
    },
    {
      "epoch": 0.8981642861993254,
      "grad_norm": 0.7201048135757446,
      "learning_rate": 2.5316114336492032e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 31915232,
      "step": 19840
    },
    {
      "epoch": 0.8983906380859684,
      "grad_norm": 1.0935115814208984,
      "learning_rate": 2.530629762019367e-05,
      "loss": 0.1775,
      "num_input_tokens_seen": 31923264,
      "step": 19845
    },
    {
      "epoch": 0.8986169899726114,
      "grad_norm": 0.9831932187080383,
      "learning_rate": 2.5296480856660364e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 31930880,
      "step": 19850
    },
    {
      "epoch": 0.8988433418592544,
      "grad_norm": 0.2908109426498413,
      "learning_rate": 2.528666404740599e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 31939424,
      "step": 19855
    },
    {
      "epoch": 0.8990696937458974,
      "grad_norm": 0.5800634622573853,
      "learning_rate": 2.527684719394442e-05,
      "loss": 0.1737,
      "num_input_tokens_seen": 31947136,
      "step": 19860
    },
    {
      "epoch": 0.8992960456325404,
      "grad_norm": 0.9102539420127869,
      "learning_rate": 2.526703029778953e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 31955328,
      "step": 19865
    },
    {
      "epoch": 0.8995223975191833,
      "grad_norm": 0.42661720514297485,
      "learning_rate": 2.5257213360455208e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 31963616,
      "step": 19870
    },
    {
      "epoch": 0.8997487494058263,
      "grad_norm": 0.4848555326461792,
      "learning_rate": 2.5247396383455353e-05,
      "loss": 0.1325,
      "num_input_tokens_seen": 31971616,
      "step": 19875
    },
    {
      "epoch": 0.8999751012924693,
      "grad_norm": 0.849837064743042,
      "learning_rate": 2.523757936830387e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 31979552,
      "step": 19880
    },
    {
      "epoch": 0.9002014531791123,
      "grad_norm": 0.6346303224563599,
      "learning_rate": 2.5227762316514662e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 31988192,
      "step": 19885
    },
    {
      "epoch": 0.9004278050657553,
      "grad_norm": 0.8879446387290955,
      "learning_rate": 2.5217945229601648e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 31995872,
      "step": 19890
    },
    {
      "epoch": 0.9006541569523981,
      "grad_norm": 0.7704745531082153,
      "learning_rate": 2.5208128109078738e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 32003776,
      "step": 19895
    },
    {
      "epoch": 0.9008805088390411,
      "grad_norm": 0.9440358281135559,
      "learning_rate": 2.5198310956459853e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 32011584,
      "step": 19900
    },
    {
      "epoch": 0.9011068607256841,
      "grad_norm": 0.3939826786518097,
      "learning_rate": 2.518849377325893e-05,
      "loss": 0.1054,
      "num_input_tokens_seen": 32019136,
      "step": 19905
    },
    {
      "epoch": 0.9013332126123271,
      "grad_norm": 0.6543247103691101,
      "learning_rate": 2.51786765609899e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 32027040,
      "step": 19910
    },
    {
      "epoch": 0.9015595644989701,
      "grad_norm": 0.4157475531101227,
      "learning_rate": 2.5168859321166694e-05,
      "loss": 0.1219,
      "num_input_tokens_seen": 32034464,
      "step": 19915
    },
    {
      "epoch": 0.9017859163856131,
      "grad_norm": 0.2671794295310974,
      "learning_rate": 2.515904205530326e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 32042336,
      "step": 19920
    },
    {
      "epoch": 0.902012268272256,
      "grad_norm": 0.5599591135978699,
      "learning_rate": 2.514922476491355e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 32051072,
      "step": 19925
    },
    {
      "epoch": 0.902238620158899,
      "grad_norm": 0.5527724623680115,
      "learning_rate": 2.51394074515115e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 32058528,
      "step": 19930
    },
    {
      "epoch": 0.902464972045542,
      "grad_norm": 0.6470173597335815,
      "learning_rate": 2.5129590116611067e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 32066496,
      "step": 19935
    },
    {
      "epoch": 0.902691323932185,
      "grad_norm": 0.529035210609436,
      "learning_rate": 2.5119772761726212e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 32074816,
      "step": 19940
    },
    {
      "epoch": 0.902917675818828,
      "grad_norm": 0.6430625915527344,
      "learning_rate": 2.5109955388370893e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 32083008,
      "step": 19945
    },
    {
      "epoch": 0.903144027705471,
      "grad_norm": 0.4586798846721649,
      "learning_rate": 2.510013799805907e-05,
      "loss": 0.1263,
      "num_input_tokens_seen": 32091296,
      "step": 19950
    },
    {
      "epoch": 0.9033703795921139,
      "grad_norm": 0.33957722783088684,
      "learning_rate": 2.5090320592304706e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 32099648,
      "step": 19955
    },
    {
      "epoch": 0.9035967314787569,
      "grad_norm": 0.6810088157653809,
      "learning_rate": 2.5080503172621777e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 32107520,
      "step": 19960
    },
    {
      "epoch": 0.9038230833653998,
      "grad_norm": 1.1311614513397217,
      "learning_rate": 2.5070685740524246e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 32115296,
      "step": 19965
    },
    {
      "epoch": 0.9040494352520428,
      "grad_norm": 0.7062297463417053,
      "learning_rate": 2.5060868297526084e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 32122784,
      "step": 19970
    },
    {
      "epoch": 0.9042757871386858,
      "grad_norm": 0.7522009611129761,
      "learning_rate": 2.5051050845141267e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 32130240,
      "step": 19975
    },
    {
      "epoch": 0.9045021390253287,
      "grad_norm": 0.35044756531715393,
      "learning_rate": 2.5041233384883765e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 32138080,
      "step": 19980
    },
    {
      "epoch": 0.9047284909119717,
      "grad_norm": 0.8871357440948486,
      "learning_rate": 2.5031415918267564e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 32145888,
      "step": 19985
    },
    {
      "epoch": 0.9049548427986147,
      "grad_norm": 0.7778136134147644,
      "learning_rate": 2.5021598446806626e-05,
      "loss": 0.1681,
      "num_input_tokens_seen": 32153248,
      "step": 19990
    },
    {
      "epoch": 0.9051811946852577,
      "grad_norm": 0.4488956332206726,
      "learning_rate": 2.5011780972014937e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 32161696,
      "step": 19995
    },
    {
      "epoch": 0.9054075465719007,
      "grad_norm": 0.5247251391410828,
      "learning_rate": 2.5001963495406478e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 32169056,
      "step": 20000
    },
    {
      "epoch": 0.9054075465719007,
      "eval_loss": 0.14731554687023163,
      "eval_runtime": 404.2137,
      "eval_samples_per_second": 97.154,
      "eval_steps_per_second": 24.289,
      "num_input_tokens_seen": 32169056,
      "step": 20000
    },
    {
      "epoch": 0.9056338984585437,
      "grad_norm": 0.5794813632965088,
      "learning_rate": 2.499214601849522e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 32177152,
      "step": 20005
    },
    {
      "epoch": 0.9058602503451866,
      "grad_norm": 0.9498506784439087,
      "learning_rate": 2.4982328542795148e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 32184992,
      "step": 20010
    },
    {
      "epoch": 0.9060866022318296,
      "grad_norm": 1.0330157279968262,
      "learning_rate": 2.497251106982024e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 32193440,
      "step": 20015
    },
    {
      "epoch": 0.9063129541184726,
      "grad_norm": 0.6332789063453674,
      "learning_rate": 2.4962693601084458e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 32201056,
      "step": 20020
    },
    {
      "epoch": 0.9065393060051156,
      "grad_norm": 0.6069503426551819,
      "learning_rate": 2.4952876138101794e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 32209152,
      "step": 20025
    },
    {
      "epoch": 0.9067656578917586,
      "grad_norm": 0.39601433277130127,
      "learning_rate": 2.4943058682386233e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 32217440,
      "step": 20030
    },
    {
      "epoch": 0.9069920097784016,
      "grad_norm": 0.5466861128807068,
      "learning_rate": 2.493324123545173e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 32225568,
      "step": 20035
    },
    {
      "epoch": 0.9072183616650444,
      "grad_norm": 0.8151348233222961,
      "learning_rate": 2.4923423798812272e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 32233472,
      "step": 20040
    },
    {
      "epoch": 0.9074447135516874,
      "grad_norm": 0.6777433753013611,
      "learning_rate": 2.4913606373981825e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 32241184,
      "step": 20045
    },
    {
      "epoch": 0.9076710654383304,
      "grad_norm": 0.3511722981929779,
      "learning_rate": 2.4903788962474357e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 32249248,
      "step": 20050
    },
    {
      "epoch": 0.9078974173249734,
      "grad_norm": 0.8100405931472778,
      "learning_rate": 2.489397156580385e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 32257344,
      "step": 20055
    },
    {
      "epoch": 0.9081237692116164,
      "grad_norm": 0.6805995106697083,
      "learning_rate": 2.4884154185484246e-05,
      "loss": 0.1297,
      "num_input_tokens_seen": 32265088,
      "step": 20060
    },
    {
      "epoch": 0.9083501210982593,
      "grad_norm": 0.7413463592529297,
      "learning_rate": 2.4874336823029526e-05,
      "loss": 0.1802,
      "num_input_tokens_seen": 32273088,
      "step": 20065
    },
    {
      "epoch": 0.9085764729849023,
      "grad_norm": 0.39273834228515625,
      "learning_rate": 2.4864519479953656e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 32281184,
      "step": 20070
    },
    {
      "epoch": 0.9088028248715453,
      "grad_norm": 1.2379051446914673,
      "learning_rate": 2.485470215777058e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 32289632,
      "step": 20075
    },
    {
      "epoch": 0.9090291767581883,
      "grad_norm": 1.3969327211380005,
      "learning_rate": 2.4844884857994258e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 32298304,
      "step": 20080
    },
    {
      "epoch": 0.9092555286448313,
      "grad_norm": 0.3681895136833191,
      "learning_rate": 2.4835067582138638e-05,
      "loss": 0.1629,
      "num_input_tokens_seen": 32306752,
      "step": 20085
    },
    {
      "epoch": 0.9094818805314743,
      "grad_norm": 0.43166354298591614,
      "learning_rate": 2.4825250331717666e-05,
      "loss": 0.1829,
      "num_input_tokens_seen": 32314688,
      "step": 20090
    },
    {
      "epoch": 0.9097082324181172,
      "grad_norm": 0.712510347366333,
      "learning_rate": 2.4815433108245298e-05,
      "loss": 0.1807,
      "num_input_tokens_seen": 32322560,
      "step": 20095
    },
    {
      "epoch": 0.9099345843047602,
      "grad_norm": 1.528594970703125,
      "learning_rate": 2.4805615913235456e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 32330176,
      "step": 20100
    },
    {
      "epoch": 0.9101609361914031,
      "grad_norm": 0.7030626535415649,
      "learning_rate": 2.479579874820208e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 32338464,
      "step": 20105
    },
    {
      "epoch": 0.9103872880780461,
      "grad_norm": 1.653502345085144,
      "learning_rate": 2.4785981614659115e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 32346560,
      "step": 20110
    },
    {
      "epoch": 0.9106136399646891,
      "grad_norm": 1.492752194404602,
      "learning_rate": 2.477616451412047e-05,
      "loss": 0.1741,
      "num_input_tokens_seen": 32355168,
      "step": 20115
    },
    {
      "epoch": 0.9108399918513321,
      "grad_norm": 0.6184459924697876,
      "learning_rate": 2.476634744810007e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 32363040,
      "step": 20120
    },
    {
      "epoch": 0.911066343737975,
      "grad_norm": 0.801308274269104,
      "learning_rate": 2.475653041811183e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 32371264,
      "step": 20125
    },
    {
      "epoch": 0.911292695624618,
      "grad_norm": 0.7047448754310608,
      "learning_rate": 2.4746713425669652e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 32379104,
      "step": 20130
    },
    {
      "epoch": 0.911519047511261,
      "grad_norm": 0.4301629960536957,
      "learning_rate": 2.4736896472287458e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 32387072,
      "step": 20135
    },
    {
      "epoch": 0.911745399397904,
      "grad_norm": 0.650004506111145,
      "learning_rate": 2.4727079559479124e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 32394464,
      "step": 20140
    },
    {
      "epoch": 0.911971751284547,
      "grad_norm": 0.5669459700584412,
      "learning_rate": 2.4717262688758557e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 32402464,
      "step": 20145
    },
    {
      "epoch": 0.9121981031711899,
      "grad_norm": 1.0705714225769043,
      "learning_rate": 2.4707445861639637e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 32410368,
      "step": 20150
    },
    {
      "epoch": 0.9124244550578329,
      "grad_norm": 0.9010360836982727,
      "learning_rate": 2.4697629079636244e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 32418272,
      "step": 20155
    },
    {
      "epoch": 0.9126508069444759,
      "grad_norm": 0.5930121541023254,
      "learning_rate": 2.4687812344262244e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 32426336,
      "step": 20160
    },
    {
      "epoch": 0.9128771588311189,
      "grad_norm": 0.6773415803909302,
      "learning_rate": 2.46779956570315e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 32434528,
      "step": 20165
    },
    {
      "epoch": 0.9131035107177619,
      "grad_norm": 0.8824792504310608,
      "learning_rate": 2.466817901945787e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 32443168,
      "step": 20170
    },
    {
      "epoch": 0.9133298626044049,
      "grad_norm": 1.0078742504119873,
      "learning_rate": 2.4658362433055217e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 32451392,
      "step": 20175
    },
    {
      "epoch": 0.9135562144910477,
      "grad_norm": 0.8831007480621338,
      "learning_rate": 2.4648545899337356e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 32459264,
      "step": 20180
    },
    {
      "epoch": 0.9137825663776907,
      "grad_norm": 0.9637875556945801,
      "learning_rate": 2.4638729419818143e-05,
      "loss": 0.1882,
      "num_input_tokens_seen": 32467104,
      "step": 20185
    },
    {
      "epoch": 0.9140089182643337,
      "grad_norm": 0.566754162311554,
      "learning_rate": 2.46289129960114e-05,
      "loss": 0.1863,
      "num_input_tokens_seen": 32474656,
      "step": 20190
    },
    {
      "epoch": 0.9142352701509767,
      "grad_norm": 0.5951328277587891,
      "learning_rate": 2.4619096629430924e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 32482464,
      "step": 20195
    },
    {
      "epoch": 0.9144616220376197,
      "grad_norm": 0.651374101638794,
      "learning_rate": 2.4609280321590543e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 32490752,
      "step": 20200
    },
    {
      "epoch": 0.9144616220376197,
      "eval_loss": 0.14699554443359375,
      "eval_runtime": 404.9497,
      "eval_samples_per_second": 96.977,
      "eval_steps_per_second": 24.245,
      "num_input_tokens_seen": 32490752,
      "step": 20200
    },
    {
      "epoch": 0.9146879739242627,
      "grad_norm": 0.7579723596572876,
      "learning_rate": 2.4599464074004037e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 32499008,
      "step": 20205
    },
    {
      "epoch": 0.9149143258109056,
      "grad_norm": 0.691066324710846,
      "learning_rate": 2.4589647888185204e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 32507136,
      "step": 20210
    },
    {
      "epoch": 0.9151406776975486,
      "grad_norm": 0.60927814245224,
      "learning_rate": 2.4579831765647836e-05,
      "loss": 0.1088,
      "num_input_tokens_seen": 32514720,
      "step": 20215
    },
    {
      "epoch": 0.9153670295841916,
      "grad_norm": 0.5604848265647888,
      "learning_rate": 2.4570015707905676e-05,
      "loss": 0.1042,
      "num_input_tokens_seen": 32522048,
      "step": 20220
    },
    {
      "epoch": 0.9155933814708346,
      "grad_norm": 0.5716456770896912,
      "learning_rate": 2.4560199716472508e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 32529792,
      "step": 20225
    },
    {
      "epoch": 0.9158197333574776,
      "grad_norm": 0.7371842861175537,
      "learning_rate": 2.455038379286207e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 32537376,
      "step": 20230
    },
    {
      "epoch": 0.9160460852441205,
      "grad_norm": 1.0041476488113403,
      "learning_rate": 2.4540567938588095e-05,
      "loss": 0.1563,
      "num_input_tokens_seen": 32545376,
      "step": 20235
    },
    {
      "epoch": 0.9162724371307635,
      "grad_norm": 0.6168292760848999,
      "learning_rate": 2.4530752155164328e-05,
      "loss": 0.1701,
      "num_input_tokens_seen": 32553600,
      "step": 20240
    },
    {
      "epoch": 0.9164987890174064,
      "grad_norm": 0.31200698018074036,
      "learning_rate": 2.4520936444104463e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 32561440,
      "step": 20245
    },
    {
      "epoch": 0.9167251409040494,
      "grad_norm": 0.9133172631263733,
      "learning_rate": 2.4511120806922218e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 32569440,
      "step": 20250
    },
    {
      "epoch": 0.9169514927906924,
      "grad_norm": 0.6521739363670349,
      "learning_rate": 2.45013052451313e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 32577984,
      "step": 20255
    },
    {
      "epoch": 0.9171778446773354,
      "grad_norm": 0.9605170488357544,
      "learning_rate": 2.4491489760245376e-05,
      "loss": 0.1582,
      "num_input_tokens_seen": 32586208,
      "step": 20260
    },
    {
      "epoch": 0.9174041965639783,
      "grad_norm": 0.891123354434967,
      "learning_rate": 2.4481674353778115e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 32593536,
      "step": 20265
    },
    {
      "epoch": 0.9176305484506213,
      "grad_norm": 0.2649250328540802,
      "learning_rate": 2.447185902724319e-05,
      "loss": 0.1269,
      "num_input_tokens_seen": 32601408,
      "step": 20270
    },
    {
      "epoch": 0.9178569003372643,
      "grad_norm": 0.5345606207847595,
      "learning_rate": 2.4462043782154233e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 32609440,
      "step": 20275
    },
    {
      "epoch": 0.9180832522239073,
      "grad_norm": 0.7986484169960022,
      "learning_rate": 2.4452228620024895e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 32617600,
      "step": 20280
    },
    {
      "epoch": 0.9183096041105503,
      "grad_norm": 0.8518936038017273,
      "learning_rate": 2.4442413542368776e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 32626400,
      "step": 20285
    },
    {
      "epoch": 0.9185359559971933,
      "grad_norm": 0.8218299746513367,
      "learning_rate": 2.4432598550699502e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 32634272,
      "step": 20290
    },
    {
      "epoch": 0.9187623078838362,
      "grad_norm": 0.4020459055900574,
      "learning_rate": 2.4422783646530663e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 32642816,
      "step": 20295
    },
    {
      "epoch": 0.9189886597704792,
      "grad_norm": 0.9608140587806702,
      "learning_rate": 2.441296883137584e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 32651680,
      "step": 20300
    },
    {
      "epoch": 0.9192150116571222,
      "grad_norm": 0.9706940054893494,
      "learning_rate": 2.4403154106748592e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 32659264,
      "step": 20305
    },
    {
      "epoch": 0.9194413635437652,
      "grad_norm": 0.671574056148529,
      "learning_rate": 2.4393339474162494e-05,
      "loss": 0.2128,
      "num_input_tokens_seen": 32667136,
      "step": 20310
    },
    {
      "epoch": 0.9196677154304082,
      "grad_norm": 0.6117256283760071,
      "learning_rate": 2.4383524935131062e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 32675232,
      "step": 20315
    },
    {
      "epoch": 0.919894067317051,
      "grad_norm": 1.4722819328308105,
      "learning_rate": 2.437371049116784e-05,
      "loss": 0.2343,
      "num_input_tokens_seen": 32683360,
      "step": 20320
    },
    {
      "epoch": 0.920120419203694,
      "grad_norm": 0.3244946002960205,
      "learning_rate": 2.436389614378632e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 32691296,
      "step": 20325
    },
    {
      "epoch": 0.920346771090337,
      "grad_norm": 0.838172197341919,
      "learning_rate": 2.435408189450002e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 32699808,
      "step": 20330
    },
    {
      "epoch": 0.92057312297698,
      "grad_norm": 0.4928062856197357,
      "learning_rate": 2.4344267744822406e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 32708224,
      "step": 20335
    },
    {
      "epoch": 0.920799474863623,
      "grad_norm": 0.4095204174518585,
      "learning_rate": 2.4334453696266944e-05,
      "loss": 0.1316,
      "num_input_tokens_seen": 32716288,
      "step": 20340
    },
    {
      "epoch": 0.921025826750266,
      "grad_norm": 0.5415951013565063,
      "learning_rate": 2.432463975034708e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 32724288,
      "step": 20345
    },
    {
      "epoch": 0.9212521786369089,
      "grad_norm": 0.5135182738304138,
      "learning_rate": 2.4314825908576265e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 32732032,
      "step": 20350
    },
    {
      "epoch": 0.9214785305235519,
      "grad_norm": 0.6937857866287231,
      "learning_rate": 2.4305012172467897e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 32740704,
      "step": 20355
    },
    {
      "epoch": 0.9217048824101949,
      "grad_norm": 0.7724782824516296,
      "learning_rate": 2.4295198543535393e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 32748192,
      "step": 20360
    },
    {
      "epoch": 0.9219312342968379,
      "grad_norm": 0.4515247642993927,
      "learning_rate": 2.4285385023292124e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 32756256,
      "step": 20365
    },
    {
      "epoch": 0.9221575861834809,
      "grad_norm": 0.4604892432689667,
      "learning_rate": 2.427557161325147e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 32764416,
      "step": 20370
    },
    {
      "epoch": 0.9223839380701239,
      "grad_norm": 0.5455062985420227,
      "learning_rate": 2.4265758314926778e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 32772608,
      "step": 20375
    },
    {
      "epoch": 0.9226102899567667,
      "grad_norm": 0.5950055718421936,
      "learning_rate": 2.4255945129831373e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 32780544,
      "step": 20380
    },
    {
      "epoch": 0.9228366418434097,
      "grad_norm": 0.6270266175270081,
      "learning_rate": 2.4246132059478578e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 32788672,
      "step": 20385
    },
    {
      "epoch": 0.9230629937300527,
      "grad_norm": 0.4607089161872864,
      "learning_rate": 2.4236319105381706e-05,
      "loss": 0.1925,
      "num_input_tokens_seen": 32796608,
      "step": 20390
    },
    {
      "epoch": 0.9232893456166957,
      "grad_norm": 0.3486397862434387,
      "learning_rate": 2.422650626905401e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 32805216,
      "step": 20395
    },
    {
      "epoch": 0.9235156975033387,
      "grad_norm": 0.46134379506111145,
      "learning_rate": 2.4216693552008785e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 32813120,
      "step": 20400
    },
    {
      "epoch": 0.9235156975033387,
      "eval_loss": 0.14716480672359467,
      "eval_runtime": 405.1152,
      "eval_samples_per_second": 96.938,
      "eval_steps_per_second": 24.235,
      "num_input_tokens_seen": 32813120,
      "step": 20400
    },
    {
      "epoch": 0.9237420493899816,
      "grad_norm": 0.3000573515892029,
      "learning_rate": 2.4206880955759247e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 32821408,
      "step": 20405
    },
    {
      "epoch": 0.9239684012766246,
      "grad_norm": 0.4464746415615082,
      "learning_rate": 2.419706848181863e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 32829600,
      "step": 20410
    },
    {
      "epoch": 0.9241947531632676,
      "grad_norm": 0.8330965638160706,
      "learning_rate": 2.4187256131700153e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 32837824,
      "step": 20415
    },
    {
      "epoch": 0.9244211050499106,
      "grad_norm": 0.37990105152130127,
      "learning_rate": 2.4177443906916985e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 32845408,
      "step": 20420
    },
    {
      "epoch": 0.9246474569365536,
      "grad_norm": 0.6933729648590088,
      "learning_rate": 2.4167631808982303e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 32852992,
      "step": 20425
    },
    {
      "epoch": 0.9248738088231966,
      "grad_norm": 0.9869456887245178,
      "learning_rate": 2.4157819839409264e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 32860320,
      "step": 20430
    },
    {
      "epoch": 0.9251001607098395,
      "grad_norm": 0.7779838442802429,
      "learning_rate": 2.414800799971098e-05,
      "loss": 0.1588,
      "num_input_tokens_seen": 32868416,
      "step": 20435
    },
    {
      "epoch": 0.9253265125964825,
      "grad_norm": 0.541577160358429,
      "learning_rate": 2.4138196291400582e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 32876160,
      "step": 20440
    },
    {
      "epoch": 0.9255528644831255,
      "grad_norm": 0.8421123027801514,
      "learning_rate": 2.412838471599114e-05,
      "loss": 0.1849,
      "num_input_tokens_seen": 32884544,
      "step": 20445
    },
    {
      "epoch": 0.9257792163697685,
      "grad_norm": 1.022661566734314,
      "learning_rate": 2.411857327499572e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 32892480,
      "step": 20450
    },
    {
      "epoch": 0.9260055682564114,
      "grad_norm": 0.469588041305542,
      "learning_rate": 2.410876196992739e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 32900640,
      "step": 20455
    },
    {
      "epoch": 0.9262319201430544,
      "grad_norm": 0.5653898119926453,
      "learning_rate": 2.4098950802299156e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 32908704,
      "step": 20460
    },
    {
      "epoch": 0.9264582720296973,
      "grad_norm": 0.8546614050865173,
      "learning_rate": 2.4089139773624027e-05,
      "loss": 0.1559,
      "num_input_tokens_seen": 32916224,
      "step": 20465
    },
    {
      "epoch": 0.9266846239163403,
      "grad_norm": 0.5292445421218872,
      "learning_rate": 2.4079328885415007e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 32924128,
      "step": 20470
    },
    {
      "epoch": 0.9269109758029833,
      "grad_norm": 0.6649433374404907,
      "learning_rate": 2.4069518139185036e-05,
      "loss": 0.1259,
      "num_input_tokens_seen": 32932832,
      "step": 20475
    },
    {
      "epoch": 0.9271373276896263,
      "grad_norm": 0.6349657773971558,
      "learning_rate": 2.405970753644706e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 32940192,
      "step": 20480
    },
    {
      "epoch": 0.9273636795762693,
      "grad_norm": 0.4694676697254181,
      "learning_rate": 2.4049897078714e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 32947904,
      "step": 20485
    },
    {
      "epoch": 0.9275900314629122,
      "grad_norm": 0.4193752706050873,
      "learning_rate": 2.404008676749874e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 32955648,
      "step": 20490
    },
    {
      "epoch": 0.9278163833495552,
      "grad_norm": 0.6305972337722778,
      "learning_rate": 2.403027660431418e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 32963584,
      "step": 20495
    },
    {
      "epoch": 0.9280427352361982,
      "grad_norm": 0.47299495339393616,
      "learning_rate": 2.402046659067314e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 32971264,
      "step": 20500
    },
    {
      "epoch": 0.9282690871228412,
      "grad_norm": 0.4683813452720642,
      "learning_rate": 2.401065672808847e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 32979744,
      "step": 20505
    },
    {
      "epoch": 0.9284954390094842,
      "grad_norm": 0.4355509281158447,
      "learning_rate": 2.400084701807296e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 32987264,
      "step": 20510
    },
    {
      "epoch": 0.9287217908961272,
      "grad_norm": 0.6270366907119751,
      "learning_rate": 2.39910374621394e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 32995328,
      "step": 20515
    },
    {
      "epoch": 0.92894814278277,
      "grad_norm": 0.6170966029167175,
      "learning_rate": 2.3981228061800544e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 33003616,
      "step": 20520
    },
    {
      "epoch": 0.929174494669413,
      "grad_norm": 0.8415407538414001,
      "learning_rate": 2.3971418818569115e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 33011680,
      "step": 20525
    },
    {
      "epoch": 0.929400846556056,
      "grad_norm": 0.370525598526001,
      "learning_rate": 2.3961609733957832e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 33019552,
      "step": 20530
    },
    {
      "epoch": 0.929627198442699,
      "grad_norm": 1.1386336088180542,
      "learning_rate": 2.395180080947939e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 33028032,
      "step": 20535
    },
    {
      "epoch": 0.929853550329342,
      "grad_norm": 1.078057050704956,
      "learning_rate": 2.394199204664642e-05,
      "loss": 0.1818,
      "num_input_tokens_seen": 33036192,
      "step": 20540
    },
    {
      "epoch": 0.930079902215985,
      "grad_norm": 0.868194043636322,
      "learning_rate": 2.3932183446971583e-05,
      "loss": 0.1755,
      "num_input_tokens_seen": 33044736,
      "step": 20545
    },
    {
      "epoch": 0.9303062541026279,
      "grad_norm": 0.4893910884857178,
      "learning_rate": 2.3922375011967473e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 33052384,
      "step": 20550
    },
    {
      "epoch": 0.9305326059892709,
      "grad_norm": 0.5690845251083374,
      "learning_rate": 2.3912566743146676e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 33060544,
      "step": 20555
    },
    {
      "epoch": 0.9307589578759139,
      "grad_norm": 0.8603192567825317,
      "learning_rate": 2.390275864202176e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 33068960,
      "step": 20560
    },
    {
      "epoch": 0.9309853097625569,
      "grad_norm": 0.6378763318061829,
      "learning_rate": 2.3892950710105243e-05,
      "loss": 0.1772,
      "num_input_tokens_seen": 33076768,
      "step": 20565
    },
    {
      "epoch": 0.9312116616491999,
      "grad_norm": 1.1214933395385742,
      "learning_rate": 2.3883142948909635e-05,
      "loss": 0.194,
      "num_input_tokens_seen": 33084896,
      "step": 20570
    },
    {
      "epoch": 0.9314380135358428,
      "grad_norm": 1.0551378726959229,
      "learning_rate": 2.3873335359947433e-05,
      "loss": 0.1105,
      "num_input_tokens_seen": 33093280,
      "step": 20575
    },
    {
      "epoch": 0.9316643654224858,
      "grad_norm": 0.422539085149765,
      "learning_rate": 2.3863527944731066e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 33101152,
      "step": 20580
    },
    {
      "epoch": 0.9318907173091288,
      "grad_norm": 0.7989985346794128,
      "learning_rate": 2.385372070477298e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 33108832,
      "step": 20585
    },
    {
      "epoch": 0.9321170691957718,
      "grad_norm": 0.7891040444374084,
      "learning_rate": 2.384391364158556e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 33117728,
      "step": 20590
    },
    {
      "epoch": 0.9323434210824147,
      "grad_norm": 0.7177416086196899,
      "learning_rate": 2.3834106756681185e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 33125600,
      "step": 20595
    },
    {
      "epoch": 0.9325697729690577,
      "grad_norm": 0.6351478099822998,
      "learning_rate": 2.3824300051572206e-05,
      "loss": 0.1153,
      "num_input_tokens_seen": 33133696,
      "step": 20600
    },
    {
      "epoch": 0.9325697729690577,
      "eval_loss": 0.14743569493293762,
      "eval_runtime": 404.069,
      "eval_samples_per_second": 97.189,
      "eval_steps_per_second": 24.298,
      "num_input_tokens_seen": 33133696,
      "step": 20600
    },
    {
      "epoch": 0.9327961248557006,
      "grad_norm": 0.33955439925193787,
      "learning_rate": 2.3814493527770923e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 33141728,
      "step": 20605
    },
    {
      "epoch": 0.9330224767423436,
      "grad_norm": 1.2876534461975098,
      "learning_rate": 2.3804687186789637e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 33149760,
      "step": 20610
    },
    {
      "epoch": 0.9332488286289866,
      "grad_norm": 0.532855212688446,
      "learning_rate": 2.379488103014062e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 33157344,
      "step": 20615
    },
    {
      "epoch": 0.9334751805156296,
      "grad_norm": 0.3515845239162445,
      "learning_rate": 2.3785075059336086e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 33165216,
      "step": 20620
    },
    {
      "epoch": 0.9337015324022726,
      "grad_norm": 0.7384480237960815,
      "learning_rate": 2.3775269275888248e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 33173312,
      "step": 20625
    },
    {
      "epoch": 0.9339278842889156,
      "grad_norm": 0.36428216099739075,
      "learning_rate": 2.3765463681309274e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 33181408,
      "step": 20630
    },
    {
      "epoch": 0.9341542361755585,
      "grad_norm": 0.7108574509620667,
      "learning_rate": 2.3755658277111313e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 33189248,
      "step": 20635
    },
    {
      "epoch": 0.9343805880622015,
      "grad_norm": 0.7289016842842102,
      "learning_rate": 2.374585306480649e-05,
      "loss": 0.1681,
      "num_input_tokens_seen": 33197248,
      "step": 20640
    },
    {
      "epoch": 0.9346069399488445,
      "grad_norm": 0.6648886203765869,
      "learning_rate": 2.3736048045906877e-05,
      "loss": 0.104,
      "num_input_tokens_seen": 33205472,
      "step": 20645
    },
    {
      "epoch": 0.9348332918354875,
      "grad_norm": 1.2280569076538086,
      "learning_rate": 2.372624322192454e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 33212960,
      "step": 20650
    },
    {
      "epoch": 0.9350596437221305,
      "grad_norm": 0.8684208393096924,
      "learning_rate": 2.3716438594371516e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 33221120,
      "step": 20655
    },
    {
      "epoch": 0.9352859956087733,
      "grad_norm": 0.33243638277053833,
      "learning_rate": 2.3706634164759784e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 33228768,
      "step": 20660
    },
    {
      "epoch": 0.9355123474954163,
      "grad_norm": 1.662540316581726,
      "learning_rate": 2.3696829934601323e-05,
      "loss": 0.1709,
      "num_input_tokens_seen": 33236576,
      "step": 20665
    },
    {
      "epoch": 0.9357386993820593,
      "grad_norm": 0.6172000169754028,
      "learning_rate": 2.3687025905408053e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 33244832,
      "step": 20670
    },
    {
      "epoch": 0.9359650512687023,
      "grad_norm": 0.4279178977012634,
      "learning_rate": 2.3677222078691886e-05,
      "loss": 0.1398,
      "num_input_tokens_seen": 33253056,
      "step": 20675
    },
    {
      "epoch": 0.9361914031553453,
      "grad_norm": 0.8302487730979919,
      "learning_rate": 2.366741845596471e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 33261408,
      "step": 20680
    },
    {
      "epoch": 0.9364177550419883,
      "grad_norm": 0.3289189636707306,
      "learning_rate": 2.3657615038738343e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 33269280,
      "step": 20685
    },
    {
      "epoch": 0.9366441069286312,
      "grad_norm": 0.574246883392334,
      "learning_rate": 2.3647811828524614e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 33277248,
      "step": 20690
    },
    {
      "epoch": 0.9368704588152742,
      "grad_norm": 0.31720098853111267,
      "learning_rate": 2.363800882683529e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 33285152,
      "step": 20695
    },
    {
      "epoch": 0.9370968107019172,
      "grad_norm": 0.3812507688999176,
      "learning_rate": 2.3628206035182125e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 33293120,
      "step": 20700
    },
    {
      "epoch": 0.9373231625885602,
      "grad_norm": 0.21223099529743195,
      "learning_rate": 2.361840345507683e-05,
      "loss": 0.124,
      "num_input_tokens_seen": 33301856,
      "step": 20705
    },
    {
      "epoch": 0.9375495144752032,
      "grad_norm": 0.2745066285133362,
      "learning_rate": 2.3608601088031073e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 33309280,
      "step": 20710
    },
    {
      "epoch": 0.9377758663618462,
      "grad_norm": 0.31746426224708557,
      "learning_rate": 2.3598798935556516e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 33316832,
      "step": 20715
    },
    {
      "epoch": 0.9380022182484891,
      "grad_norm": 0.41609859466552734,
      "learning_rate": 2.3588996999164784e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 33324896,
      "step": 20720
    },
    {
      "epoch": 0.9382285701351321,
      "grad_norm": 0.4888496696949005,
      "learning_rate": 2.3579195280367434e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 33332512,
      "step": 20725
    },
    {
      "epoch": 0.938454922021775,
      "grad_norm": 0.641794741153717,
      "learning_rate": 2.356939378067603e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 33340352,
      "step": 20730
    },
    {
      "epoch": 0.938681273908418,
      "grad_norm": 0.5894063115119934,
      "learning_rate": 2.3559592501602092e-05,
      "loss": 0.1539,
      "num_input_tokens_seen": 33348608,
      "step": 20735
    },
    {
      "epoch": 0.938907625795061,
      "grad_norm": 0.6771718859672546,
      "learning_rate": 2.3549791444657076e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 33357312,
      "step": 20740
    },
    {
      "epoch": 0.9391339776817039,
      "grad_norm": 0.6787290573120117,
      "learning_rate": 2.353999061135246e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 33365344,
      "step": 20745
    },
    {
      "epoch": 0.9393603295683469,
      "grad_norm": 0.5424860119819641,
      "learning_rate": 2.3530190003199626e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 33373408,
      "step": 20750
    },
    {
      "epoch": 0.9395866814549899,
      "grad_norm": 0.5738372802734375,
      "learning_rate": 2.3520389621709965e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 33381088,
      "step": 20755
    },
    {
      "epoch": 0.9398130333416329,
      "grad_norm": 0.7162361741065979,
      "learning_rate": 2.351058946839483e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 33388672,
      "step": 20760
    },
    {
      "epoch": 0.9400393852282759,
      "grad_norm": 0.6749934554100037,
      "learning_rate": 2.350078954476551e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 33396736,
      "step": 20765
    },
    {
      "epoch": 0.9402657371149189,
      "grad_norm": 0.6801218390464783,
      "learning_rate": 2.3490989852333272e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 33404896,
      "step": 20770
    },
    {
      "epoch": 0.9404920890015618,
      "grad_norm": 0.45693162083625793,
      "learning_rate": 2.3481190392609377e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 33412576,
      "step": 20775
    },
    {
      "epoch": 0.9407184408882048,
      "grad_norm": 0.44984522461891174,
      "learning_rate": 2.3471391167105e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 33420448,
      "step": 20780
    },
    {
      "epoch": 0.9409447927748478,
      "grad_norm": 0.5284605026245117,
      "learning_rate": 2.3461592177331325e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 33428416,
      "step": 20785
    },
    {
      "epoch": 0.9411711446614908,
      "grad_norm": 0.3406399190425873,
      "learning_rate": 2.345179342479946e-05,
      "loss": 0.1174,
      "num_input_tokens_seen": 33436224,
      "step": 20790
    },
    {
      "epoch": 0.9413974965481338,
      "grad_norm": 0.5631494522094727,
      "learning_rate": 2.3441994911020503e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 33444288,
      "step": 20795
    },
    {
      "epoch": 0.9416238484347768,
      "grad_norm": 0.68208909034729,
      "learning_rate": 2.3432196637505522e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 33452320,
      "step": 20800
    },
    {
      "epoch": 0.9416238484347768,
      "eval_loss": 0.14772388339042664,
      "eval_runtime": 404.7726,
      "eval_samples_per_second": 97.02,
      "eval_steps_per_second": 24.256,
      "num_input_tokens_seen": 33452320,
      "step": 20800
    },
    {
      "epoch": 0.9418502003214196,
      "grad_norm": 0.9517109394073486,
      "learning_rate": 2.3422398605765515e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 33460384,
      "step": 20805
    },
    {
      "epoch": 0.9420765522080626,
      "grad_norm": 0.5146124362945557,
      "learning_rate": 2.3412600817311462e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 33467680,
      "step": 20810
    },
    {
      "epoch": 0.9423029040947056,
      "grad_norm": 0.4334716498851776,
      "learning_rate": 2.3402803273654326e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 33475648,
      "step": 20815
    },
    {
      "epoch": 0.9425292559813486,
      "grad_norm": 0.36337727308273315,
      "learning_rate": 2.3393005976304983e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 33483392,
      "step": 20820
    },
    {
      "epoch": 0.9427556078679916,
      "grad_norm": 0.4007863402366638,
      "learning_rate": 2.338320892677432e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 33491040,
      "step": 20825
    },
    {
      "epoch": 0.9429819597546345,
      "grad_norm": 0.5004162788391113,
      "learning_rate": 2.3373412126573155e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 33499072,
      "step": 20830
    },
    {
      "epoch": 0.9432083116412775,
      "grad_norm": 0.32714512944221497,
      "learning_rate": 2.3363615577212285e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 33507008,
      "step": 20835
    },
    {
      "epoch": 0.9434346635279205,
      "grad_norm": 0.5528114438056946,
      "learning_rate": 2.3353819280202455e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 33515328,
      "step": 20840
    },
    {
      "epoch": 0.9436610154145635,
      "grad_norm": 0.3956262767314911,
      "learning_rate": 2.334402323705438e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 33523392,
      "step": 20845
    },
    {
      "epoch": 0.9438873673012065,
      "grad_norm": 0.8673408031463623,
      "learning_rate": 2.3334227449278725e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 33532704,
      "step": 20850
    },
    {
      "epoch": 0.9441137191878495,
      "grad_norm": 0.7981648445129395,
      "learning_rate": 2.3324431918386143e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 33540736,
      "step": 20855
    },
    {
      "epoch": 0.9443400710744924,
      "grad_norm": 0.412535160779953,
      "learning_rate": 2.3314636645887207e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 33548320,
      "step": 20860
    },
    {
      "epoch": 0.9445664229611354,
      "grad_norm": 0.3240596652030945,
      "learning_rate": 2.3304841633292487e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 33556448,
      "step": 20865
    },
    {
      "epoch": 0.9447927748477783,
      "grad_norm": 0.6441272497177124,
      "learning_rate": 2.329504688211248e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 33563872,
      "step": 20870
    },
    {
      "epoch": 0.9450191267344213,
      "grad_norm": 1.193649172782898,
      "learning_rate": 2.3285252393857677e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 33572096,
      "step": 20875
    },
    {
      "epoch": 0.9452454786210643,
      "grad_norm": 1.1092649698257446,
      "learning_rate": 2.327545817003851e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 33580576,
      "step": 20880
    },
    {
      "epoch": 0.9454718305077073,
      "grad_norm": 0.48816558718681335,
      "learning_rate": 2.326566421216535e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 33589056,
      "step": 20885
    },
    {
      "epoch": 0.9456981823943502,
      "grad_norm": 0.44651854038238525,
      "learning_rate": 2.3255870521748565e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 33597152,
      "step": 20890
    },
    {
      "epoch": 0.9459245342809932,
      "grad_norm": 0.8830904364585876,
      "learning_rate": 2.3246077100298474e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 33605088,
      "step": 20895
    },
    {
      "epoch": 0.9461508861676362,
      "grad_norm": 0.5036613941192627,
      "learning_rate": 2.3236283949325328e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 33613184,
      "step": 20900
    },
    {
      "epoch": 0.9463772380542792,
      "grad_norm": 0.8627535700798035,
      "learning_rate": 2.3226491070339368e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 33620960,
      "step": 20905
    },
    {
      "epoch": 0.9466035899409222,
      "grad_norm": 0.41623246669769287,
      "learning_rate": 2.3216698464850762e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 33628960,
      "step": 20910
    },
    {
      "epoch": 0.9468299418275651,
      "grad_norm": 1.0804251432418823,
      "learning_rate": 2.320690613436967e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 33636640,
      "step": 20915
    },
    {
      "epoch": 0.9470562937142081,
      "grad_norm": 0.6384842395782471,
      "learning_rate": 2.3197114080406192e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 33644448,
      "step": 20920
    },
    {
      "epoch": 0.9472826456008511,
      "grad_norm": 0.38829782605171204,
      "learning_rate": 2.3187322304470365e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 33652960,
      "step": 20925
    },
    {
      "epoch": 0.9475089974874941,
      "grad_norm": 0.7650966644287109,
      "learning_rate": 2.3177530808072222e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 33661088,
      "step": 20930
    },
    {
      "epoch": 0.9477353493741371,
      "grad_norm": 0.9120515584945679,
      "learning_rate": 2.316773959272174e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 33669376,
      "step": 20935
    },
    {
      "epoch": 0.94796170126078,
      "grad_norm": 0.6839418411254883,
      "learning_rate": 2.3157948659928823e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 33677248,
      "step": 20940
    },
    {
      "epoch": 0.9481880531474229,
      "grad_norm": 0.8969916701316833,
      "learning_rate": 2.3148158011203388e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 33685536,
      "step": 20945
    },
    {
      "epoch": 0.9484144050340659,
      "grad_norm": 0.9514481425285339,
      "learning_rate": 2.3138367648055253e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 33693824,
      "step": 20950
    },
    {
      "epoch": 0.9486407569207089,
      "grad_norm": 0.6852895617485046,
      "learning_rate": 2.312857757199422e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 33701760,
      "step": 20955
    },
    {
      "epoch": 0.9488671088073519,
      "grad_norm": 1.2549229860305786,
      "learning_rate": 2.3118787784530048e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 33711136,
      "step": 20960
    },
    {
      "epoch": 0.9490934606939949,
      "grad_norm": 0.4570097029209137,
      "learning_rate": 2.310899828717243e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 33719424,
      "step": 20965
    },
    {
      "epoch": 0.9493198125806379,
      "grad_norm": 1.4695779085159302,
      "learning_rate": 2.309920908143104e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 33727072,
      "step": 20970
    },
    {
      "epoch": 0.9495461644672808,
      "grad_norm": 0.4376735985279083,
      "learning_rate": 2.308942016881551e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 33735168,
      "step": 20975
    },
    {
      "epoch": 0.9497725163539238,
      "grad_norm": 0.8671814799308777,
      "learning_rate": 2.307963155083539e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 33742976,
      "step": 20980
    },
    {
      "epoch": 0.9499988682405668,
      "grad_norm": 0.32649603486061096,
      "learning_rate": 2.306984322900022e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 33751424,
      "step": 20985
    },
    {
      "epoch": 0.9502252201272098,
      "grad_norm": 0.4725677967071533,
      "learning_rate": 2.3060055204819482e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 33759648,
      "step": 20990
    },
    {
      "epoch": 0.9504515720138528,
      "grad_norm": 0.5738493800163269,
      "learning_rate": 2.3050267479802604e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 33767840,
      "step": 20995
    },
    {
      "epoch": 0.9506779239004957,
      "grad_norm": 0.5247540473937988,
      "learning_rate": 2.304048005545899e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 33776032,
      "step": 21000
    },
    {
      "epoch": 0.9506779239004957,
      "eval_loss": 0.1467217206954956,
      "eval_runtime": 404.4216,
      "eval_samples_per_second": 97.104,
      "eval_steps_per_second": 24.277,
      "num_input_tokens_seen": 33776032,
      "step": 21000
    },
    {
      "epoch": 0.9509042757871387,
      "grad_norm": 1.0384083986282349,
      "learning_rate": 2.3030692933297972e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 33784192,
      "step": 21005
    },
    {
      "epoch": 0.9511306276737816,
      "grad_norm": 0.8642218708992004,
      "learning_rate": 2.3020906114828843e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 33793280,
      "step": 21010
    },
    {
      "epoch": 0.9513569795604246,
      "grad_norm": 0.888202965259552,
      "learning_rate": 2.301111960156088e-05,
      "loss": 0.1068,
      "num_input_tokens_seen": 33801312,
      "step": 21015
    },
    {
      "epoch": 0.9515833314470676,
      "grad_norm": 1.1586802005767822,
      "learning_rate": 2.300133339500326e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 33809696,
      "step": 21020
    },
    {
      "epoch": 0.9518096833337106,
      "grad_norm": 0.5374242663383484,
      "learning_rate": 2.2991547496665148e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 33817664,
      "step": 21025
    },
    {
      "epoch": 0.9520360352203535,
      "grad_norm": 0.4817052185535431,
      "learning_rate": 2.298176190805565e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 33826016,
      "step": 21030
    },
    {
      "epoch": 0.9522623871069965,
      "grad_norm": 0.5837623476982117,
      "learning_rate": 2.2971976630683826e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 33834080,
      "step": 21035
    },
    {
      "epoch": 0.9524887389936395,
      "grad_norm": 1.1161922216415405,
      "learning_rate": 2.29621916660587e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 33842368,
      "step": 21040
    },
    {
      "epoch": 0.9527150908802825,
      "grad_norm": 0.3716656565666199,
      "learning_rate": 2.295240701568922e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 33850208,
      "step": 21045
    },
    {
      "epoch": 0.9529414427669255,
      "grad_norm": 0.4044111669063568,
      "learning_rate": 2.2942622681084312e-05,
      "loss": 0.1768,
      "num_input_tokens_seen": 33857632,
      "step": 21050
    },
    {
      "epoch": 0.9531677946535685,
      "grad_norm": 0.2860806882381439,
      "learning_rate": 2.293283866375284e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 33865504,
      "step": 21055
    },
    {
      "epoch": 0.9533941465402114,
      "grad_norm": 0.4762797951698303,
      "learning_rate": 2.2923054965203627e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 33873888,
      "step": 21060
    },
    {
      "epoch": 0.9536204984268544,
      "grad_norm": 0.5108733773231506,
      "learning_rate": 2.2913271586945443e-05,
      "loss": 0.1269,
      "num_input_tokens_seen": 33882272,
      "step": 21065
    },
    {
      "epoch": 0.9538468503134974,
      "grad_norm": 0.460204541683197,
      "learning_rate": 2.290348853048699e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 33890432,
      "step": 21070
    },
    {
      "epoch": 0.9540732022001404,
      "grad_norm": 0.639724612236023,
      "learning_rate": 2.2893705797336956e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 33898208,
      "step": 21075
    },
    {
      "epoch": 0.9542995540867834,
      "grad_norm": 0.7550246715545654,
      "learning_rate": 2.288392338900397e-05,
      "loss": 0.1779,
      "num_input_tokens_seen": 33907104,
      "step": 21080
    },
    {
      "epoch": 0.9545259059734262,
      "grad_norm": 0.9684120416641235,
      "learning_rate": 2.2874141306996576e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 33915392,
      "step": 21085
    },
    {
      "epoch": 0.9547522578600692,
      "grad_norm": 1.088255763053894,
      "learning_rate": 2.2864359552823312e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 33923104,
      "step": 21090
    },
    {
      "epoch": 0.9549786097467122,
      "grad_norm": 0.5803527235984802,
      "learning_rate": 2.2854578127992648e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 33930752,
      "step": 21095
    },
    {
      "epoch": 0.9552049616333552,
      "grad_norm": 0.32493552565574646,
      "learning_rate": 2.2844797034012988e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 33938592,
      "step": 21100
    },
    {
      "epoch": 0.9554313135199982,
      "grad_norm": 0.9716498255729675,
      "learning_rate": 2.2835016272392722e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 33946720,
      "step": 21105
    },
    {
      "epoch": 0.9556576654066412,
      "grad_norm": 0.6099035143852234,
      "learning_rate": 2.2825235844640142e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 33955104,
      "step": 21110
    },
    {
      "epoch": 0.9558840172932841,
      "grad_norm": 0.49580737948417664,
      "learning_rate": 2.2815455752263522e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 33963104,
      "step": 21115
    },
    {
      "epoch": 0.9561103691799271,
      "grad_norm": 1.0799627304077148,
      "learning_rate": 2.2805675996771092e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 33970784,
      "step": 21120
    },
    {
      "epoch": 0.9563367210665701,
      "grad_norm": 0.7434067726135254,
      "learning_rate": 2.2795896579670987e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 33978624,
      "step": 21125
    },
    {
      "epoch": 0.9565630729532131,
      "grad_norm": 0.45763787627220154,
      "learning_rate": 2.2786117502471337e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 33986176,
      "step": 21130
    },
    {
      "epoch": 0.9567894248398561,
      "grad_norm": 0.7715013027191162,
      "learning_rate": 2.2776338766680185e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 33994048,
      "step": 21135
    },
    {
      "epoch": 0.9570157767264991,
      "grad_norm": 0.7404031157493591,
      "learning_rate": 2.2766560373805533e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 34001632,
      "step": 21140
    },
    {
      "epoch": 0.957242128613142,
      "grad_norm": 0.534755289554596,
      "learning_rate": 2.2756782325355353e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 34010592,
      "step": 21145
    },
    {
      "epoch": 0.957468480499785,
      "grad_norm": 0.7284175753593445,
      "learning_rate": 2.2747004622837514e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 34018464,
      "step": 21150
    },
    {
      "epoch": 0.9576948323864279,
      "grad_norm": 0.7348290681838989,
      "learning_rate": 2.2737227267759878e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 34026848,
      "step": 21155
    },
    {
      "epoch": 0.9579211842730709,
      "grad_norm": 0.7383341193199158,
      "learning_rate": 2.272745026163024e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 34035200,
      "step": 21160
    },
    {
      "epoch": 0.9581475361597139,
      "grad_norm": 0.36828672885894775,
      "learning_rate": 2.271767360595633e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 34043264,
      "step": 21165
    },
    {
      "epoch": 0.9583738880463568,
      "grad_norm": 0.29938098788261414,
      "learning_rate": 2.270789730224583e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 34050912,
      "step": 21170
    },
    {
      "epoch": 0.9586002399329998,
      "grad_norm": 0.4864792227745056,
      "learning_rate": 2.2698121352006367e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 34059808,
      "step": 21175
    },
    {
      "epoch": 0.9588265918196428,
      "grad_norm": 0.3016667664051056,
      "learning_rate": 2.2688345756745517e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 34067648,
      "step": 21180
    },
    {
      "epoch": 0.9590529437062858,
      "grad_norm": 0.47097131609916687,
      "learning_rate": 2.267857051797081e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 34075552,
      "step": 21185
    },
    {
      "epoch": 0.9592792955929288,
      "grad_norm": 0.4095723628997803,
      "learning_rate": 2.2668795637189695e-05,
      "loss": 0.1359,
      "num_input_tokens_seen": 34083648,
      "step": 21190
    },
    {
      "epoch": 0.9595056474795718,
      "grad_norm": 0.6511597037315369,
      "learning_rate": 2.2659021115909586e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 34091488,
      "step": 21195
    },
    {
      "epoch": 0.9597319993662147,
      "grad_norm": 1.0199558734893799,
      "learning_rate": 2.2649246955637847e-05,
      "loss": 0.1469,
      "num_input_tokens_seen": 34099680,
      "step": 21200
    },
    {
      "epoch": 0.9597319993662147,
      "eval_loss": 0.14681631326675415,
      "eval_runtime": 404.5727,
      "eval_samples_per_second": 97.068,
      "eval_steps_per_second": 24.268,
      "num_input_tokens_seen": 34099680,
      "step": 21200
    },
    {
      "epoch": 0.9599583512528577,
      "grad_norm": 0.3114546537399292,
      "learning_rate": 2.2639473157881766e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 34107808,
      "step": 21205
    },
    {
      "epoch": 0.9601847031395007,
      "grad_norm": 0.8526507616043091,
      "learning_rate": 2.2629699724148594e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 34116608,
      "step": 21210
    },
    {
      "epoch": 0.9604110550261437,
      "grad_norm": 0.46163448691368103,
      "learning_rate": 2.26199266559455e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 34123808,
      "step": 21215
    },
    {
      "epoch": 0.9606374069127867,
      "grad_norm": 0.7757999897003174,
      "learning_rate": 2.2610153954779625e-05,
      "loss": 0.1344,
      "num_input_tokens_seen": 34131840,
      "step": 21220
    },
    {
      "epoch": 0.9608637587994296,
      "grad_norm": 0.6695551872253418,
      "learning_rate": 2.2600381622158056e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 34140960,
      "step": 21225
    },
    {
      "epoch": 0.9610901106860725,
      "grad_norm": 0.7211385369300842,
      "learning_rate": 2.2590609659587783e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 34148768,
      "step": 21230
    },
    {
      "epoch": 0.9613164625727155,
      "grad_norm": 0.8846794962882996,
      "learning_rate": 2.2580838068575787e-05,
      "loss": 0.1304,
      "num_input_tokens_seen": 34157376,
      "step": 21235
    },
    {
      "epoch": 0.9615428144593585,
      "grad_norm": 0.4121907651424408,
      "learning_rate": 2.257106685062896e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 34165440,
      "step": 21240
    },
    {
      "epoch": 0.9617691663460015,
      "grad_norm": 0.5707591772079468,
      "learning_rate": 2.256129600725415e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 34173696,
      "step": 21245
    },
    {
      "epoch": 0.9619955182326445,
      "grad_norm": 0.2991304099559784,
      "learning_rate": 2.2551525539958145e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 34181888,
      "step": 21250
    },
    {
      "epoch": 0.9622218701192874,
      "grad_norm": 0.8453980088233948,
      "learning_rate": 2.2541755450247663e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 34190048,
      "step": 21255
    },
    {
      "epoch": 0.9624482220059304,
      "grad_norm": 0.5072763562202454,
      "learning_rate": 2.2531985739629382e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 34198240,
      "step": 21260
    },
    {
      "epoch": 0.9626745738925734,
      "grad_norm": 0.8644075989723206,
      "learning_rate": 2.2522216409609924e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 34206400,
      "step": 21265
    },
    {
      "epoch": 0.9629009257792164,
      "grad_norm": 0.7571772336959839,
      "learning_rate": 2.2512447461695826e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 34216384,
      "step": 21270
    },
    {
      "epoch": 0.9631272776658594,
      "grad_norm": 0.44349271059036255,
      "learning_rate": 2.2502678897393593e-05,
      "loss": 0.1803,
      "num_input_tokens_seen": 34224768,
      "step": 21275
    },
    {
      "epoch": 0.9633536295525024,
      "grad_norm": 0.38617512583732605,
      "learning_rate": 2.2492910718209665e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 34233152,
      "step": 21280
    },
    {
      "epoch": 0.9635799814391453,
      "grad_norm": 1.2027889490127563,
      "learning_rate": 2.2483142925650398e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 34241056,
      "step": 21285
    },
    {
      "epoch": 0.9638063333257882,
      "grad_norm": 0.6414772272109985,
      "learning_rate": 2.247337552122213e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 34249056,
      "step": 21290
    },
    {
      "epoch": 0.9640326852124312,
      "grad_norm": 0.7437915205955505,
      "learning_rate": 2.24636085064311e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 34257536,
      "step": 21295
    },
    {
      "epoch": 0.9642590370990742,
      "grad_norm": 0.6050093173980713,
      "learning_rate": 2.245384188278351e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 34265440,
      "step": 21300
    },
    {
      "epoch": 0.9644853889857172,
      "grad_norm": 0.7259789705276489,
      "learning_rate": 2.2444075651785513e-05,
      "loss": 0.1651,
      "num_input_tokens_seen": 34273472,
      "step": 21305
    },
    {
      "epoch": 0.9647117408723602,
      "grad_norm": 1.0868041515350342,
      "learning_rate": 2.243430981494316e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 34281280,
      "step": 21310
    },
    {
      "epoch": 0.9649380927590031,
      "grad_norm": 0.9170043468475342,
      "learning_rate": 2.2424544373762475e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 34288768,
      "step": 21315
    },
    {
      "epoch": 0.9651644446456461,
      "grad_norm": 0.737011194229126,
      "learning_rate": 2.2414779329749418e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 34296672,
      "step": 21320
    },
    {
      "epoch": 0.9653907965322891,
      "grad_norm": 1.4911240339279175,
      "learning_rate": 2.2405014684409873e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 34304320,
      "step": 21325
    },
    {
      "epoch": 0.9656171484189321,
      "grad_norm": 1.1348037719726562,
      "learning_rate": 2.239525043924968e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 34312064,
      "step": 21330
    },
    {
      "epoch": 0.9658435003055751,
      "grad_norm": 0.5090815424919128,
      "learning_rate": 2.2385486595774592e-05,
      "loss": 0.1731,
      "num_input_tokens_seen": 34320320,
      "step": 21335
    },
    {
      "epoch": 0.966069852192218,
      "grad_norm": 0.5937796831130981,
      "learning_rate": 2.237572315549033e-05,
      "loss": 0.1492,
      "num_input_tokens_seen": 34327712,
      "step": 21340
    },
    {
      "epoch": 0.966296204078861,
      "grad_norm": 0.8319137692451477,
      "learning_rate": 2.2365960119902545e-05,
      "loss": 0.172,
      "num_input_tokens_seen": 34335424,
      "step": 21345
    },
    {
      "epoch": 0.966522555965504,
      "grad_norm": 0.24394835531711578,
      "learning_rate": 2.2356197490516806e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 34343872,
      "step": 21350
    },
    {
      "epoch": 0.966748907852147,
      "grad_norm": 0.5031188726425171,
      "learning_rate": 2.234643526883863e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 34352288,
      "step": 21355
    },
    {
      "epoch": 0.96697525973879,
      "grad_norm": 0.6448420882225037,
      "learning_rate": 2.2336673456373497e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 34360224,
      "step": 21360
    },
    {
      "epoch": 0.9672016116254329,
      "grad_norm": 0.7298598289489746,
      "learning_rate": 2.2326912054626772e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 34369056,
      "step": 21365
    },
    {
      "epoch": 0.9674279635120758,
      "grad_norm": 0.4675354063510895,
      "learning_rate": 2.2317151065103813e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 34376992,
      "step": 21370
    },
    {
      "epoch": 0.9676543153987188,
      "grad_norm": 0.9577482342720032,
      "learning_rate": 2.2307390489309865e-05,
      "loss": 0.1801,
      "num_input_tokens_seen": 34384864,
      "step": 21375
    },
    {
      "epoch": 0.9678806672853618,
      "grad_norm": 1.2407058477401733,
      "learning_rate": 2.2297630328750146e-05,
      "loss": 0.1227,
      "num_input_tokens_seen": 34392512,
      "step": 21380
    },
    {
      "epoch": 0.9681070191720048,
      "grad_norm": 0.837054431438446,
      "learning_rate": 2.228787058492979e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 34400416,
      "step": 21385
    },
    {
      "epoch": 0.9683333710586478,
      "grad_norm": 0.4058256149291992,
      "learning_rate": 2.2278111259353875e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 34409088,
      "step": 21390
    },
    {
      "epoch": 0.9685597229452908,
      "grad_norm": 0.8055698871612549,
      "learning_rate": 2.2268352353527395e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 34416928,
      "step": 21395
    },
    {
      "epoch": 0.9687860748319337,
      "grad_norm": 0.7506183385848999,
      "learning_rate": 2.225859386895533e-05,
      "loss": 0.1443,
      "num_input_tokens_seen": 34424928,
      "step": 21400
    },
    {
      "epoch": 0.9687860748319337,
      "eval_loss": 0.14742179214954376,
      "eval_runtime": 405.5529,
      "eval_samples_per_second": 96.833,
      "eval_steps_per_second": 24.209,
      "num_input_tokens_seen": 34424928,
      "step": 21400
    },
    {
      "epoch": 0.9690124267185767,
      "grad_norm": 0.35628658533096313,
      "learning_rate": 2.2248835807142525e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 34433056,
      "step": 21405
    },
    {
      "epoch": 0.9692387786052197,
      "grad_norm": 0.7823324203491211,
      "learning_rate": 2.2239078169593826e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 34440736,
      "step": 21410
    },
    {
      "epoch": 0.9694651304918627,
      "grad_norm": 0.4209827184677124,
      "learning_rate": 2.222932095781396e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 34448576,
      "step": 21415
    },
    {
      "epoch": 0.9696914823785057,
      "grad_norm": 0.2957499325275421,
      "learning_rate": 2.221956417330762e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 34457056,
      "step": 21420
    },
    {
      "epoch": 0.9699178342651485,
      "grad_norm": 0.5399700999259949,
      "learning_rate": 2.2209807817579438e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 34465056,
      "step": 21425
    },
    {
      "epoch": 0.9701441861517915,
      "grad_norm": 0.7295712232589722,
      "learning_rate": 2.220005189213394e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 34472928,
      "step": 21430
    },
    {
      "epoch": 0.9703705380384345,
      "grad_norm": 0.5127325057983398,
      "learning_rate": 2.2190296398475624e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 34481120,
      "step": 21435
    },
    {
      "epoch": 0.9705968899250775,
      "grad_norm": 0.6996890306472778,
      "learning_rate": 2.2180541338108926e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 34489024,
      "step": 21440
    },
    {
      "epoch": 0.9708232418117205,
      "grad_norm": 0.6371135115623474,
      "learning_rate": 2.2170786712538176e-05,
      "loss": 0.1885,
      "num_input_tokens_seen": 34496736,
      "step": 21445
    },
    {
      "epoch": 0.9710495936983635,
      "grad_norm": 0.8652201890945435,
      "learning_rate": 2.216103252326768e-05,
      "loss": 0.1072,
      "num_input_tokens_seen": 34505344,
      "step": 21450
    },
    {
      "epoch": 0.9712759455850064,
      "grad_norm": 1.2073721885681152,
      "learning_rate": 2.2151278771801635e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 34514368,
      "step": 21455
    },
    {
      "epoch": 0.9715022974716494,
      "grad_norm": 0.5087610483169556,
      "learning_rate": 2.21415254596442e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 34522240,
      "step": 21460
    },
    {
      "epoch": 0.9717286493582924,
      "grad_norm": 0.2728574573993683,
      "learning_rate": 2.213177258829947e-05,
      "loss": 0.1016,
      "num_input_tokens_seen": 34530144,
      "step": 21465
    },
    {
      "epoch": 0.9719550012449354,
      "grad_norm": 1.0507198572158813,
      "learning_rate": 2.2122020159271445e-05,
      "loss": 0.1681,
      "num_input_tokens_seen": 34537952,
      "step": 21470
    },
    {
      "epoch": 0.9721813531315784,
      "grad_norm": 0.4754534363746643,
      "learning_rate": 2.2112268174064075e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 34547456,
      "step": 21475
    },
    {
      "epoch": 0.9724077050182214,
      "grad_norm": 0.41472333669662476,
      "learning_rate": 2.2102516634181253e-05,
      "loss": 0.1134,
      "num_input_tokens_seen": 34555552,
      "step": 21480
    },
    {
      "epoch": 0.9726340569048643,
      "grad_norm": 0.607211709022522,
      "learning_rate": 2.209276554112677e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 34563808,
      "step": 21485
    },
    {
      "epoch": 0.9728604087915073,
      "grad_norm": 0.659324586391449,
      "learning_rate": 2.2083014896404384e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 34571936,
      "step": 21490
    },
    {
      "epoch": 0.9730867606781503,
      "grad_norm": 0.8323732614517212,
      "learning_rate": 2.207326470151775e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 34579616,
      "step": 21495
    },
    {
      "epoch": 0.9733131125647932,
      "grad_norm": 0.3758734166622162,
      "learning_rate": 2.2063514957970477e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 34588192,
      "step": 21500
    },
    {
      "epoch": 0.9735394644514362,
      "grad_norm": 0.7078402042388916,
      "learning_rate": 2.205376566726611e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 34596352,
      "step": 21505
    },
    {
      "epoch": 0.9737658163380791,
      "grad_norm": 0.6224841475486755,
      "learning_rate": 2.204401683090809e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 34605024,
      "step": 21510
    },
    {
      "epoch": 0.9739921682247221,
      "grad_norm": 0.6526482701301575,
      "learning_rate": 2.203426845039982e-05,
      "loss": 0.0965,
      "num_input_tokens_seen": 34613056,
      "step": 21515
    },
    {
      "epoch": 0.9742185201113651,
      "grad_norm": 0.7392874360084534,
      "learning_rate": 2.202452052724464e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 34621472,
      "step": 21520
    },
    {
      "epoch": 0.9744448719980081,
      "grad_norm": 0.5261018872261047,
      "learning_rate": 2.2014773062945777e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 34629152,
      "step": 21525
    },
    {
      "epoch": 0.9746712238846511,
      "grad_norm": 0.5156830549240112,
      "learning_rate": 2.2005026059006427e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 34637248,
      "step": 21530
    },
    {
      "epoch": 0.9748975757712941,
      "grad_norm": 0.3378513753414154,
      "learning_rate": 2.1995279516929695e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 34645056,
      "step": 21535
    },
    {
      "epoch": 0.975123927657937,
      "grad_norm": 0.43840616941452026,
      "learning_rate": 2.1985533438218613e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 34652704,
      "step": 21540
    },
    {
      "epoch": 0.97535027954458,
      "grad_norm": 1.1235262155532837,
      "learning_rate": 2.197578782437617e-05,
      "loss": 0.1751,
      "num_input_tokens_seen": 34660512,
      "step": 21545
    },
    {
      "epoch": 0.975576631431223,
      "grad_norm": 1.0094653367996216,
      "learning_rate": 2.196604267690524e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 34668480,
      "step": 21550
    },
    {
      "epoch": 0.975802983317866,
      "grad_norm": 0.5754347443580627,
      "learning_rate": 2.195629799730865e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 34676384,
      "step": 21555
    },
    {
      "epoch": 0.976029335204509,
      "grad_norm": 1.0997267961502075,
      "learning_rate": 2.1946553787089173e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 34685632,
      "step": 21560
    },
    {
      "epoch": 0.976255687091152,
      "grad_norm": 0.8908462524414062,
      "learning_rate": 2.193681004774947e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 34693568,
      "step": 21565
    },
    {
      "epoch": 0.9764820389777948,
      "grad_norm": 0.5477364659309387,
      "learning_rate": 2.1927066780792154e-05,
      "loss": 0.1096,
      "num_input_tokens_seen": 34701440,
      "step": 21570
    },
    {
      "epoch": 0.9767083908644378,
      "grad_norm": 0.3309652507305145,
      "learning_rate": 2.191732398771975e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 34709088,
      "step": 21575
    },
    {
      "epoch": 0.9769347427510808,
      "grad_norm": 0.7172495126724243,
      "learning_rate": 2.1907581670034725e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 34717120,
      "step": 21580
    },
    {
      "epoch": 0.9771610946377238,
      "grad_norm": 0.5241601467132568,
      "learning_rate": 2.189783982923948e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 34724512,
      "step": 21585
    },
    {
      "epoch": 0.9773874465243668,
      "grad_norm": 0.4000072181224823,
      "learning_rate": 2.1888098466836303e-05,
      "loss": 0.0984,
      "num_input_tokens_seen": 34732128,
      "step": 21590
    },
    {
      "epoch": 0.9776137984110097,
      "grad_norm": 1.0238677263259888,
      "learning_rate": 2.1878357584327457e-05,
      "loss": 0.1302,
      "num_input_tokens_seen": 34740320,
      "step": 21595
    },
    {
      "epoch": 0.9778401502976527,
      "grad_norm": 0.7279909253120422,
      "learning_rate": 2.1868617183215103e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 34748544,
      "step": 21600
    },
    {
      "epoch": 0.9778401502976527,
      "eval_loss": 0.14642953872680664,
      "eval_runtime": 405.2006,
      "eval_samples_per_second": 96.917,
      "eval_steps_per_second": 24.23,
      "num_input_tokens_seen": 34748544,
      "step": 21600
    },
    {
      "epoch": 0.9780665021842957,
      "grad_norm": 0.7957366704940796,
      "learning_rate": 2.1858877265001327e-05,
      "loss": 0.1687,
      "num_input_tokens_seen": 34756512,
      "step": 21605
    },
    {
      "epoch": 0.9782928540709387,
      "grad_norm": 0.671747088432312,
      "learning_rate": 2.184913783118816e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 34765152,
      "step": 21610
    },
    {
      "epoch": 0.9785192059575817,
      "grad_norm": 0.32763779163360596,
      "learning_rate": 2.1839398883277522e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 34773440,
      "step": 21615
    },
    {
      "epoch": 0.9787455578442247,
      "grad_norm": 0.5091896653175354,
      "learning_rate": 2.182966042277129e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 34781472,
      "step": 21620
    },
    {
      "epoch": 0.9789719097308676,
      "grad_norm": 0.35154134035110474,
      "learning_rate": 2.181992245117128e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 34788928,
      "step": 21625
    },
    {
      "epoch": 0.9791982616175106,
      "grad_norm": 0.41648995876312256,
      "learning_rate": 2.181018496997918e-05,
      "loss": 0.1685,
      "num_input_tokens_seen": 34796960,
      "step": 21630
    },
    {
      "epoch": 0.9794246135041536,
      "grad_norm": 0.5635231137275696,
      "learning_rate": 2.1800447980696648e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 34805088,
      "step": 21635
    },
    {
      "epoch": 0.9796509653907965,
      "grad_norm": 0.8610899448394775,
      "learning_rate": 2.1790711484825248e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 34812928,
      "step": 21640
    },
    {
      "epoch": 0.9798773172774395,
      "grad_norm": 0.9845597743988037,
      "learning_rate": 2.178097548386646e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 34820576,
      "step": 21645
    },
    {
      "epoch": 0.9801036691640825,
      "grad_norm": 0.30229732394218445,
      "learning_rate": 2.1771239979321712e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 34828512,
      "step": 21650
    },
    {
      "epoch": 0.9803300210507254,
      "grad_norm": 0.3351421058177948,
      "learning_rate": 2.1761504972692327e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 34836480,
      "step": 21655
    },
    {
      "epoch": 0.9805563729373684,
      "grad_norm": 0.3946510851383209,
      "learning_rate": 2.1751770465479572e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 34844576,
      "step": 21660
    },
    {
      "epoch": 0.9807827248240114,
      "grad_norm": 0.656377911567688,
      "learning_rate": 2.174203645918464e-05,
      "loss": 0.1607,
      "num_input_tokens_seen": 34852256,
      "step": 21665
    },
    {
      "epoch": 0.9810090767106544,
      "grad_norm": 0.841738224029541,
      "learning_rate": 2.1732302955308624e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 34860288,
      "step": 21670
    },
    {
      "epoch": 0.9812354285972974,
      "grad_norm": 0.6701703071594238,
      "learning_rate": 2.172256995535255e-05,
      "loss": 0.1717,
      "num_input_tokens_seen": 34867840,
      "step": 21675
    },
    {
      "epoch": 0.9814617804839403,
      "grad_norm": 0.565049409866333,
      "learning_rate": 2.171283746081739e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 34876064,
      "step": 21680
    },
    {
      "epoch": 0.9816881323705833,
      "grad_norm": 0.29756978154182434,
      "learning_rate": 2.1703105473203988e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 34884096,
      "step": 21685
    },
    {
      "epoch": 0.9819144842572263,
      "grad_norm": 0.3200281262397766,
      "learning_rate": 2.1693373994013168e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 34891968,
      "step": 21690
    },
    {
      "epoch": 0.9821408361438693,
      "grad_norm": 0.507027268409729,
      "learning_rate": 2.168364302474562e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 34899392,
      "step": 21695
    },
    {
      "epoch": 0.9823671880305123,
      "grad_norm": 0.9642364978790283,
      "learning_rate": 2.167391256690199e-05,
      "loss": 0.1551,
      "num_input_tokens_seen": 34907808,
      "step": 21700
    },
    {
      "epoch": 0.9825935399171553,
      "grad_norm": 1.3909258842468262,
      "learning_rate": 2.1664182621982855e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 34915680,
      "step": 21705
    },
    {
      "epoch": 0.9828198918037981,
      "grad_norm": 0.7641861438751221,
      "learning_rate": 2.1654453191488673e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 34924672,
      "step": 21710
    },
    {
      "epoch": 0.9830462436904411,
      "grad_norm": 1.1904218196868896,
      "learning_rate": 2.1644724276919846e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 34932576,
      "step": 21715
    },
    {
      "epoch": 0.9832725955770841,
      "grad_norm": 0.7086918950080872,
      "learning_rate": 2.1634995879776715e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 34940544,
      "step": 21720
    },
    {
      "epoch": 0.9834989474637271,
      "grad_norm": 0.5887830257415771,
      "learning_rate": 2.162526800155949e-05,
      "loss": 0.1746,
      "num_input_tokens_seen": 34948480,
      "step": 21725
    },
    {
      "epoch": 0.9837252993503701,
      "grad_norm": 0.7483288049697876,
      "learning_rate": 2.1615540643768363e-05,
      "loss": 0.1769,
      "num_input_tokens_seen": 34957440,
      "step": 21730
    },
    {
      "epoch": 0.9839516512370131,
      "grad_norm": 0.696598470211029,
      "learning_rate": 2.160581380790339e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 34964928,
      "step": 21735
    },
    {
      "epoch": 0.984178003123656,
      "grad_norm": 0.472002238035202,
      "learning_rate": 2.1596087495464586e-05,
      "loss": 0.1778,
      "num_input_tokens_seen": 34972960,
      "step": 21740
    },
    {
      "epoch": 0.984404355010299,
      "grad_norm": 0.973523736000061,
      "learning_rate": 2.1586361707951866e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 34981312,
      "step": 21745
    },
    {
      "epoch": 0.984630706896942,
      "grad_norm": 0.6008665561676025,
      "learning_rate": 2.157663644686507e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 34989888,
      "step": 21750
    },
    {
      "epoch": 0.984857058783585,
      "grad_norm": 0.8100993633270264,
      "learning_rate": 2.156691171370396e-05,
      "loss": 0.1841,
      "num_input_tokens_seen": 34998528,
      "step": 21755
    },
    {
      "epoch": 0.985083410670228,
      "grad_norm": 0.4559473395347595,
      "learning_rate": 2.1557187509968195e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 35006656,
      "step": 21760
    },
    {
      "epoch": 0.9853097625568709,
      "grad_norm": 0.8672731518745422,
      "learning_rate": 2.1547463837157382e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 35014656,
      "step": 21765
    },
    {
      "epoch": 0.9855361144435139,
      "grad_norm": 1.6916871070861816,
      "learning_rate": 2.1537740696771045e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 35022624,
      "step": 21770
    },
    {
      "epoch": 0.9857624663301569,
      "grad_norm": 0.4023900330066681,
      "learning_rate": 2.1528018090308587e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 35030656,
      "step": 21775
    },
    {
      "epoch": 0.9859888182167998,
      "grad_norm": 0.7863446474075317,
      "learning_rate": 2.151829601926938e-05,
      "loss": 0.1496,
      "num_input_tokens_seen": 35038752,
      "step": 21780
    },
    {
      "epoch": 0.9862151701034428,
      "grad_norm": 0.5243304371833801,
      "learning_rate": 2.1508574485152684e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 35046752,
      "step": 21785
    },
    {
      "epoch": 0.9864415219900858,
      "grad_norm": 0.3872745931148529,
      "learning_rate": 2.1498853489457667e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 35054624,
      "step": 21790
    },
    {
      "epoch": 0.9866678738767287,
      "grad_norm": 1.0428311824798584,
      "learning_rate": 2.1489133033683455e-05,
      "loss": 0.1214,
      "num_input_tokens_seen": 35063296,
      "step": 21795
    },
    {
      "epoch": 0.9868942257633717,
      "grad_norm": 0.5948784351348877,
      "learning_rate": 2.1479413119329038e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 35071104,
      "step": 21800
    },
    {
      "epoch": 0.9868942257633717,
      "eval_loss": 0.14628663659095764,
      "eval_runtime": 404.4206,
      "eval_samples_per_second": 97.104,
      "eval_steps_per_second": 24.277,
      "num_input_tokens_seen": 35071104,
      "step": 21800
    },
    {
      "epoch": 0.9871205776500147,
      "grad_norm": 0.5001301169395447,
      "learning_rate": 2.1469693747893355e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 35078944,
      "step": 21805
    },
    {
      "epoch": 0.9873469295366577,
      "grad_norm": 0.7869800329208374,
      "learning_rate": 2.1459974920875274e-05,
      "loss": 0.1538,
      "num_input_tokens_seen": 35086944,
      "step": 21810
    },
    {
      "epoch": 0.9875732814233007,
      "grad_norm": 0.5860437750816345,
      "learning_rate": 2.145025663977354e-05,
      "loss": 0.184,
      "num_input_tokens_seen": 35095072,
      "step": 21815
    },
    {
      "epoch": 0.9877996333099437,
      "grad_norm": 0.42348939180374146,
      "learning_rate": 2.1440538906086844e-05,
      "loss": 0.1117,
      "num_input_tokens_seen": 35103200,
      "step": 21820
    },
    {
      "epoch": 0.9880259851965866,
      "grad_norm": 0.4911308288574219,
      "learning_rate": 2.1430821721313782e-05,
      "loss": 0.1275,
      "num_input_tokens_seen": 35112128,
      "step": 21825
    },
    {
      "epoch": 0.9882523370832296,
      "grad_norm": 0.5981737971305847,
      "learning_rate": 2.142110508695286e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 35120224,
      "step": 21830
    },
    {
      "epoch": 0.9884786889698726,
      "grad_norm": 0.718132495880127,
      "learning_rate": 2.1411389004502515e-05,
      "loss": 0.107,
      "num_input_tokens_seen": 35128096,
      "step": 21835
    },
    {
      "epoch": 0.9887050408565156,
      "grad_norm": 0.5183557271957397,
      "learning_rate": 2.140167347546107e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 35135904,
      "step": 21840
    },
    {
      "epoch": 0.9889313927431586,
      "grad_norm": 1.0426011085510254,
      "learning_rate": 2.1391958501326793e-05,
      "loss": 0.1592,
      "num_input_tokens_seen": 35144000,
      "step": 21845
    },
    {
      "epoch": 0.9891577446298014,
      "grad_norm": 0.8149588108062744,
      "learning_rate": 2.1382244083597873e-05,
      "loss": 0.1648,
      "num_input_tokens_seen": 35152608,
      "step": 21850
    },
    {
      "epoch": 0.9893840965164444,
      "grad_norm": 0.834895670413971,
      "learning_rate": 2.137253022377237e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 35160288,
      "step": 21855
    },
    {
      "epoch": 0.9896104484030874,
      "grad_norm": 0.6612374782562256,
      "learning_rate": 2.136281692334829e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 35168000,
      "step": 21860
    },
    {
      "epoch": 0.9898368002897304,
      "grad_norm": 0.34305626153945923,
      "learning_rate": 2.135310418382356e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 35175520,
      "step": 21865
    },
    {
      "epoch": 0.9900631521763734,
      "grad_norm": 0.35640305280685425,
      "learning_rate": 2.134339200669598e-05,
      "loss": 0.1323,
      "num_input_tokens_seen": 35183552,
      "step": 21870
    },
    {
      "epoch": 0.9902895040630164,
      "grad_norm": 0.5749332308769226,
      "learning_rate": 2.133368039346331e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 35190912,
      "step": 21875
    },
    {
      "epoch": 0.9905158559496593,
      "grad_norm": 0.6286994814872742,
      "learning_rate": 2.1323969345623195e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 35198752,
      "step": 21880
    },
    {
      "epoch": 0.9907422078363023,
      "grad_norm": 0.4313356280326843,
      "learning_rate": 2.1314258864673207e-05,
      "loss": 0.1329,
      "num_input_tokens_seen": 35206784,
      "step": 21885
    },
    {
      "epoch": 0.9909685597229453,
      "grad_norm": 0.4898377060890198,
      "learning_rate": 2.130454895211082e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 35214816,
      "step": 21890
    },
    {
      "epoch": 0.9911949116095883,
      "grad_norm": 1.3260656595230103,
      "learning_rate": 2.129483960943342e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 35222560,
      "step": 21895
    },
    {
      "epoch": 0.9914212634962313,
      "grad_norm": 0.3288341760635376,
      "learning_rate": 2.128513083813831e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 35230208,
      "step": 21900
    },
    {
      "epoch": 0.9916476153828743,
      "grad_norm": 0.4387655556201935,
      "learning_rate": 2.1275422639722724e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 35237888,
      "step": 21905
    },
    {
      "epoch": 0.9918739672695172,
      "grad_norm": 0.31812387704849243,
      "learning_rate": 2.126571501568376e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 35246016,
      "step": 21910
    },
    {
      "epoch": 0.9921003191561601,
      "grad_norm": 0.4444160759449005,
      "learning_rate": 2.1256007967518478e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 35253856,
      "step": 21915
    },
    {
      "epoch": 0.9923266710428031,
      "grad_norm": 0.5450275540351868,
      "learning_rate": 2.124630149672381e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 35261696,
      "step": 21920
    },
    {
      "epoch": 0.9925530229294461,
      "grad_norm": 0.5747339129447937,
      "learning_rate": 2.1236595604796624e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 35269696,
      "step": 21925
    },
    {
      "epoch": 0.9927793748160891,
      "grad_norm": 0.4281969964504242,
      "learning_rate": 2.1226890293233693e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 35277344,
      "step": 21930
    },
    {
      "epoch": 0.993005726702732,
      "grad_norm": 0.32246026396751404,
      "learning_rate": 2.1217185563531694e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 35285568,
      "step": 21935
    },
    {
      "epoch": 0.993232078589375,
      "grad_norm": 0.35350534319877625,
      "learning_rate": 2.120748141718721e-05,
      "loss": 0.1523,
      "num_input_tokens_seen": 35293376,
      "step": 21940
    },
    {
      "epoch": 0.993458430476018,
      "grad_norm": 0.9897263050079346,
      "learning_rate": 2.1197777855696765e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 35301408,
      "step": 21945
    },
    {
      "epoch": 0.993684782362661,
      "grad_norm": 1.3480656147003174,
      "learning_rate": 2.1188074880556746e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 35308960,
      "step": 21950
    },
    {
      "epoch": 0.993911134249304,
      "grad_norm": 0.7467007040977478,
      "learning_rate": 2.1178372493263495e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 35316704,
      "step": 21955
    },
    {
      "epoch": 0.994137486135947,
      "grad_norm": 1.0666919946670532,
      "learning_rate": 2.116867069531322e-05,
      "loss": 0.1824,
      "num_input_tokens_seen": 35324480,
      "step": 21960
    },
    {
      "epoch": 0.9943638380225899,
      "grad_norm": 0.5492631196975708,
      "learning_rate": 2.1158969488202073e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 35332256,
      "step": 21965
    },
    {
      "epoch": 0.9945901899092329,
      "grad_norm": 0.4359816312789917,
      "learning_rate": 2.114926887342611e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 35339872,
      "step": 21970
    },
    {
      "epoch": 0.9948165417958759,
      "grad_norm": 0.4706695079803467,
      "learning_rate": 2.113956885248127e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 35347936,
      "step": 21975
    },
    {
      "epoch": 0.9950428936825189,
      "grad_norm": 0.7778224349021912,
      "learning_rate": 2.112986942686342e-05,
      "loss": 0.1668,
      "num_input_tokens_seen": 35355680,
      "step": 21980
    },
    {
      "epoch": 0.9952692455691619,
      "grad_norm": 0.2736826539039612,
      "learning_rate": 2.112017059806835e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 35363936,
      "step": 21985
    },
    {
      "epoch": 0.9954955974558048,
      "grad_norm": 0.9405673146247864,
      "learning_rate": 2.1110472367591724e-05,
      "loss": 0.1797,
      "num_input_tokens_seen": 35372864,
      "step": 21990
    },
    {
      "epoch": 0.9957219493424477,
      "grad_norm": 0.5594533681869507,
      "learning_rate": 2.1100774736929145e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 35381088,
      "step": 21995
    },
    {
      "epoch": 0.9959483012290907,
      "grad_norm": 0.5126089453697205,
      "learning_rate": 2.10910777075761e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 35388960,
      "step": 22000
    },
    {
      "epoch": 0.9959483012290907,
      "eval_loss": 0.14724768698215485,
      "eval_runtime": 404.1359,
      "eval_samples_per_second": 97.173,
      "eval_steps_per_second": 24.294,
      "num_input_tokens_seen": 35388960,
      "step": 22000
    },
    {
      "epoch": 0.9961746531157337,
      "grad_norm": 0.3883945047855377,
      "learning_rate": 2.108138128102799e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 35396416,
      "step": 22005
    },
    {
      "epoch": 0.9964010050023767,
      "grad_norm": 0.6406975984573364,
      "learning_rate": 2.107168545878014e-05,
      "loss": 0.0859,
      "num_input_tokens_seen": 35405248,
      "step": 22010
    },
    {
      "epoch": 0.9966273568890197,
      "grad_norm": 0.7342075109481812,
      "learning_rate": 2.106199024232775e-05,
      "loss": 0.2133,
      "num_input_tokens_seen": 35412960,
      "step": 22015
    },
    {
      "epoch": 0.9968537087756626,
      "grad_norm": 0.6450848579406738,
      "learning_rate": 2.105229563316595e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 35420928,
      "step": 22020
    },
    {
      "epoch": 0.9970800606623056,
      "grad_norm": 0.6699177622795105,
      "learning_rate": 2.1042601632789784e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 35428544,
      "step": 22025
    },
    {
      "epoch": 0.9973064125489486,
      "grad_norm": 0.6565819978713989,
      "learning_rate": 2.103290824269417e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 35436512,
      "step": 22030
    },
    {
      "epoch": 0.9975327644355916,
      "grad_norm": 0.6820458173751831,
      "learning_rate": 2.1023215464373965e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 35445088,
      "step": 22035
    },
    {
      "epoch": 0.9977591163222346,
      "grad_norm": 1.4062700271606445,
      "learning_rate": 2.1013523299323908e-05,
      "loss": 0.1321,
      "num_input_tokens_seen": 35453216,
      "step": 22040
    },
    {
      "epoch": 0.9979854682088776,
      "grad_norm": 0.9340587854385376,
      "learning_rate": 2.1003831749038654e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 35461728,
      "step": 22045
    },
    {
      "epoch": 0.9982118200955205,
      "grad_norm": 0.49030712246894836,
      "learning_rate": 2.099414081501277e-05,
      "loss": 0.1277,
      "num_input_tokens_seen": 35469312,
      "step": 22050
    },
    {
      "epoch": 0.9984381719821634,
      "grad_norm": 0.6500596404075623,
      "learning_rate": 2.09844504987407e-05,
      "loss": 0.1339,
      "num_input_tokens_seen": 35476960,
      "step": 22055
    },
    {
      "epoch": 0.9986645238688064,
      "grad_norm": 0.43859949707984924,
      "learning_rate": 2.097476080171683e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 35485472,
      "step": 22060
    },
    {
      "epoch": 0.9988908757554494,
      "grad_norm": 1.1707359552383423,
      "learning_rate": 2.0965071725435436e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 35493856,
      "step": 22065
    },
    {
      "epoch": 0.9991172276420924,
      "grad_norm": 0.43483635783195496,
      "learning_rate": 2.0955383271390684e-05,
      "loss": 0.1647,
      "num_input_tokens_seen": 35502144,
      "step": 22070
    },
    {
      "epoch": 0.9993435795287354,
      "grad_norm": 0.8042669296264648,
      "learning_rate": 2.094569544107666e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 35510368,
      "step": 22075
    },
    {
      "epoch": 0.9995699314153783,
      "grad_norm": 0.6133525371551514,
      "learning_rate": 2.093600823598735e-05,
      "loss": 0.1221,
      "num_input_tokens_seen": 35518432,
      "step": 22080
    },
    {
      "epoch": 0.9997962833020213,
      "grad_norm": 0.3897414207458496,
      "learning_rate": 2.092632165761663e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 35527648,
      "step": 22085
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.59228253364563,
      "learning_rate": 2.091663570745832e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 35534744,
      "step": 22090
    },
    {
      "epoch": 1.000226351886643,
      "grad_norm": 0.7246991991996765,
      "learning_rate": 2.0906950387006086e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 35542424,
      "step": 22095
    },
    {
      "epoch": 1.000452703773286,
      "grad_norm": 0.3671351373195648,
      "learning_rate": 2.0897265697753543e-05,
      "loss": 0.1866,
      "num_input_tokens_seen": 35550840,
      "step": 22100
    },
    {
      "epoch": 1.000679055659929,
      "grad_norm": 1.001458764076233,
      "learning_rate": 2.088758164119419e-05,
      "loss": 0.1574,
      "num_input_tokens_seen": 35558936,
      "step": 22105
    },
    {
      "epoch": 1.000905407546572,
      "grad_norm": 0.6690317988395691,
      "learning_rate": 2.0877898218821428e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 35566936,
      "step": 22110
    },
    {
      "epoch": 1.001131759433215,
      "grad_norm": 0.4136376976966858,
      "learning_rate": 2.0868215432128565e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 35574584,
      "step": 22115
    },
    {
      "epoch": 1.0013581113198577,
      "grad_norm": 0.5280313491821289,
      "learning_rate": 2.0858533282608796e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 35582776,
      "step": 22120
    },
    {
      "epoch": 1.0015844632065007,
      "grad_norm": 0.6594773530960083,
      "learning_rate": 2.084885177175524e-05,
      "loss": 0.168,
      "num_input_tokens_seen": 35590808,
      "step": 22125
    },
    {
      "epoch": 1.0018108150931437,
      "grad_norm": 0.5655260682106018,
      "learning_rate": 2.0839170901060917e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 35598712,
      "step": 22130
    },
    {
      "epoch": 1.0020371669797867,
      "grad_norm": 1.1777039766311646,
      "learning_rate": 2.082949067201872e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 35606424,
      "step": 22135
    },
    {
      "epoch": 1.0022635188664297,
      "grad_norm": 0.3619535565376282,
      "learning_rate": 2.0819811086121475e-05,
      "loss": 0.1216,
      "num_input_tokens_seen": 35614296,
      "step": 22140
    },
    {
      "epoch": 1.0024898707530727,
      "grad_norm": 0.5687800049781799,
      "learning_rate": 2.08101321448619e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 35622808,
      "step": 22145
    },
    {
      "epoch": 1.0027162226397157,
      "grad_norm": 0.5525425672531128,
      "learning_rate": 2.080045384973259e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 35630360,
      "step": 22150
    },
    {
      "epoch": 1.0029425745263587,
      "grad_norm": 1.0807933807373047,
      "learning_rate": 2.0790776202226082e-05,
      "loss": 0.1861,
      "num_input_tokens_seen": 35637976,
      "step": 22155
    },
    {
      "epoch": 1.0031689264130017,
      "grad_norm": 0.6731915473937988,
      "learning_rate": 2.078109920383477e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 35645784,
      "step": 22160
    },
    {
      "epoch": 1.0033952782996447,
      "grad_norm": 0.37582528591156006,
      "learning_rate": 2.0771422856050978e-05,
      "loss": 0.094,
      "num_input_tokens_seen": 35654008,
      "step": 22165
    },
    {
      "epoch": 1.0036216301862877,
      "grad_norm": 0.5175727009773254,
      "learning_rate": 2.076174716036693e-05,
      "loss": 0.1285,
      "num_input_tokens_seen": 35662520,
      "step": 22170
    },
    {
      "epoch": 1.0038479820729305,
      "grad_norm": 0.5397278666496277,
      "learning_rate": 2.075207211827472e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 35670808,
      "step": 22175
    },
    {
      "epoch": 1.0040743339595735,
      "grad_norm": 0.5905038118362427,
      "learning_rate": 2.074239773126638e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 35679640,
      "step": 22180
    },
    {
      "epoch": 1.0043006858462165,
      "grad_norm": 0.5272305011749268,
      "learning_rate": 2.073272400083382e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 35687704,
      "step": 22185
    },
    {
      "epoch": 1.0045270377328595,
      "grad_norm": 1.029419183731079,
      "learning_rate": 2.072305092846883e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 35695352,
      "step": 22190
    },
    {
      "epoch": 1.0047533896195024,
      "grad_norm": 1.6804214715957642,
      "learning_rate": 2.0713378515663152e-05,
      "loss": 0.1906,
      "num_input_tokens_seen": 35703672,
      "step": 22195
    },
    {
      "epoch": 1.0049797415061454,
      "grad_norm": 0.618106484413147,
      "learning_rate": 2.070370676390836e-05,
      "loss": 0.0938,
      "num_input_tokens_seen": 35712792,
      "step": 22200
    },
    {
      "epoch": 1.0049797415061454,
      "eval_loss": 0.14610929787158966,
      "eval_runtime": 404.6682,
      "eval_samples_per_second": 97.045,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 35712792,
      "step": 22200
    },
    {
      "epoch": 1.0052060933927884,
      "grad_norm": 0.7963980436325073,
      "learning_rate": 2.0694035674695974e-05,
      "loss": 0.1333,
      "num_input_tokens_seen": 35721080,
      "step": 22205
    },
    {
      "epoch": 1.0054324452794314,
      "grad_norm": 0.4730178117752075,
      "learning_rate": 2.0684365249517416e-05,
      "loss": 0.1611,
      "num_input_tokens_seen": 35728984,
      "step": 22210
    },
    {
      "epoch": 1.0056587971660744,
      "grad_norm": 0.3033260703086853,
      "learning_rate": 2.067469548986396e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 35737080,
      "step": 22215
    },
    {
      "epoch": 1.0058851490527174,
      "grad_norm": 0.4462639093399048,
      "learning_rate": 2.066502639722681e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 35744664,
      "step": 22220
    },
    {
      "epoch": 1.0061115009393604,
      "grad_norm": 1.3863070011138916,
      "learning_rate": 2.065535797309708e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 35752344,
      "step": 22225
    },
    {
      "epoch": 1.0063378528260034,
      "grad_norm": 0.5460673570632935,
      "learning_rate": 2.0645690218965736e-05,
      "loss": 0.1301,
      "num_input_tokens_seen": 35760376,
      "step": 22230
    },
    {
      "epoch": 1.0065642047126462,
      "grad_norm": 0.33663249015808105,
      "learning_rate": 2.063602313632369e-05,
      "loss": 0.1675,
      "num_input_tokens_seen": 35767896,
      "step": 22235
    },
    {
      "epoch": 1.0067905565992892,
      "grad_norm": 0.5107477307319641,
      "learning_rate": 2.0626356726661704e-05,
      "loss": 0.1076,
      "num_input_tokens_seen": 35775896,
      "step": 22240
    },
    {
      "epoch": 1.0070169084859322,
      "grad_norm": 0.592994213104248,
      "learning_rate": 2.0616690991470477e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 35783832,
      "step": 22245
    },
    {
      "epoch": 1.0072432603725752,
      "grad_norm": 0.9737153649330139,
      "learning_rate": 2.0607025932240595e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 35791704,
      "step": 22250
    },
    {
      "epoch": 1.0074696122592182,
      "grad_norm": 1.275292992591858,
      "learning_rate": 2.059736155046251e-05,
      "loss": 0.1236,
      "num_input_tokens_seen": 35799896,
      "step": 22255
    },
    {
      "epoch": 1.0076959641458612,
      "grad_norm": 0.7905048727989197,
      "learning_rate": 2.0587697847626603e-05,
      "loss": 0.1205,
      "num_input_tokens_seen": 35807480,
      "step": 22260
    },
    {
      "epoch": 1.0079223160325042,
      "grad_norm": 0.6926992535591125,
      "learning_rate": 2.057803482522314e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 35815544,
      "step": 22265
    },
    {
      "epoch": 1.0081486679191471,
      "grad_norm": 0.7246147990226746,
      "learning_rate": 2.056837248474227e-05,
      "loss": 0.1024,
      "num_input_tokens_seen": 35823640,
      "step": 22270
    },
    {
      "epoch": 1.0083750198057901,
      "grad_norm": 0.5311410427093506,
      "learning_rate": 2.0558710827674064e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 35831576,
      "step": 22275
    },
    {
      "epoch": 1.0086013716924331,
      "grad_norm": 0.482759028673172,
      "learning_rate": 2.054904985550845e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 35839832,
      "step": 22280
    },
    {
      "epoch": 1.0088277235790761,
      "grad_norm": 1.0146968364715576,
      "learning_rate": 2.0539389569735287e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 35848120,
      "step": 22285
    },
    {
      "epoch": 1.009054075465719,
      "grad_norm": 0.7210794687271118,
      "learning_rate": 2.052972997184431e-05,
      "loss": 0.0944,
      "num_input_tokens_seen": 35855608,
      "step": 22290
    },
    {
      "epoch": 1.009280427352362,
      "grad_norm": 0.7250410914421082,
      "learning_rate": 2.0520071063325146e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 35864568,
      "step": 22295
    },
    {
      "epoch": 1.009506779239005,
      "grad_norm": 0.44630953669548035,
      "learning_rate": 2.051041284566732e-05,
      "loss": 0.1552,
      "num_input_tokens_seen": 35872376,
      "step": 22300
    },
    {
      "epoch": 1.0097331311256479,
      "grad_norm": 0.5498194098472595,
      "learning_rate": 2.050075532036026e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 35880408,
      "step": 22305
    },
    {
      "epoch": 1.0099594830122909,
      "grad_norm": 1.3637536764144897,
      "learning_rate": 2.0491098488893264e-05,
      "loss": 0.1505,
      "num_input_tokens_seen": 35888440,
      "step": 22310
    },
    {
      "epoch": 1.0101858348989339,
      "grad_norm": 0.8143194913864136,
      "learning_rate": 2.0481442352755546e-05,
      "loss": 0.1393,
      "num_input_tokens_seen": 35896824,
      "step": 22315
    },
    {
      "epoch": 1.0104121867855769,
      "grad_norm": 0.9527003765106201,
      "learning_rate": 2.0471786913436198e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 35905272,
      "step": 22320
    },
    {
      "epoch": 1.0106385386722199,
      "grad_norm": 0.854306161403656,
      "learning_rate": 2.0462132172424218e-05,
      "loss": 0.109,
      "num_input_tokens_seen": 35913400,
      "step": 22325
    },
    {
      "epoch": 1.0108648905588629,
      "grad_norm": 1.0294007062911987,
      "learning_rate": 2.0452478131208484e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 35921720,
      "step": 22330
    },
    {
      "epoch": 1.0110912424455059,
      "grad_norm": 0.43070247769355774,
      "learning_rate": 2.0442824791277765e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 35929080,
      "step": 22335
    },
    {
      "epoch": 1.0113175943321489,
      "grad_norm": 0.5064215064048767,
      "learning_rate": 2.0433172154120727e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 35937624,
      "step": 22340
    },
    {
      "epoch": 1.0115439462187918,
      "grad_norm": 1.0778859853744507,
      "learning_rate": 2.0423520221225947e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 35945816,
      "step": 22345
    },
    {
      "epoch": 1.0117702981054346,
      "grad_norm": 0.30774641036987305,
      "learning_rate": 2.0413868994081848e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 35954072,
      "step": 22350
    },
    {
      "epoch": 1.0119966499920776,
      "grad_norm": 1.1406898498535156,
      "learning_rate": 2.0404218474176795e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 35962200,
      "step": 22355
    },
    {
      "epoch": 1.0122230018787206,
      "grad_norm": 0.7223683595657349,
      "learning_rate": 2.0394568662999002e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 35970360,
      "step": 22360
    },
    {
      "epoch": 1.0124493537653636,
      "grad_norm": 0.6479731798171997,
      "learning_rate": 2.0384919562036593e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 35978488,
      "step": 22365
    },
    {
      "epoch": 1.0126757056520066,
      "grad_norm": 0.35958608984947205,
      "learning_rate": 2.0375271172777593e-05,
      "loss": 0.1307,
      "num_input_tokens_seen": 35986264,
      "step": 22370
    },
    {
      "epoch": 1.0129020575386496,
      "grad_norm": 0.5916337370872498,
      "learning_rate": 2.0365623496709885e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 35994392,
      "step": 22375
    },
    {
      "epoch": 1.0131284094252926,
      "grad_norm": 0.7496772408485413,
      "learning_rate": 2.0355976535321283e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 36002424,
      "step": 22380
    },
    {
      "epoch": 1.0133547613119356,
      "grad_norm": 0.3710784614086151,
      "learning_rate": 2.034633029009945e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 36010392,
      "step": 22385
    },
    {
      "epoch": 1.0135811131985786,
      "grad_norm": 0.5623818039894104,
      "learning_rate": 2.0336684762531972e-05,
      "loss": 0.1087,
      "num_input_tokens_seen": 36018360,
      "step": 22390
    },
    {
      "epoch": 1.0138074650852216,
      "grad_norm": 0.712187647819519,
      "learning_rate": 2.032703995410631e-05,
      "loss": 0.1454,
      "num_input_tokens_seen": 36025624,
      "step": 22395
    },
    {
      "epoch": 1.0140338169718646,
      "grad_norm": 0.5702168345451355,
      "learning_rate": 2.031739586630981e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 36032952,
      "step": 22400
    },
    {
      "epoch": 1.0140338169718646,
      "eval_loss": 0.1469167321920395,
      "eval_runtime": 404.8018,
      "eval_samples_per_second": 97.013,
      "eval_steps_per_second": 24.254,
      "num_input_tokens_seen": 36032952,
      "step": 22400
    },
    {
      "epoch": 1.0142601688585073,
      "grad_norm": 0.8095841407775879,
      "learning_rate": 2.0307752500629707e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 36040760,
      "step": 22405
    },
    {
      "epoch": 1.0144865207451503,
      "grad_norm": 0.4689668118953705,
      "learning_rate": 2.0298109858553144e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 36048888,
      "step": 22410
    },
    {
      "epoch": 1.0147128726317933,
      "grad_norm": 0.6549654006958008,
      "learning_rate": 2.028846794156712e-05,
      "loss": 0.1255,
      "num_input_tokens_seen": 36056184,
      "step": 22415
    },
    {
      "epoch": 1.0149392245184363,
      "grad_norm": 0.6196061968803406,
      "learning_rate": 2.027882675115856e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 36063896,
      "step": 22420
    },
    {
      "epoch": 1.0151655764050793,
      "grad_norm": 0.7859944701194763,
      "learning_rate": 2.026918628881423e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 36072120,
      "step": 22425
    },
    {
      "epoch": 1.0153919282917223,
      "grad_norm": 0.4409676194190979,
      "learning_rate": 2.0259546556020833e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 36079768,
      "step": 22430
    },
    {
      "epoch": 1.0156182801783653,
      "grad_norm": 0.6577968597412109,
      "learning_rate": 2.024990755426493e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 36087576,
      "step": 22435
    },
    {
      "epoch": 1.0158446320650083,
      "grad_norm": 0.37484341859817505,
      "learning_rate": 2.0240269285032975e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 36096056,
      "step": 22440
    },
    {
      "epoch": 1.0160709839516513,
      "grad_norm": 0.738286018371582,
      "learning_rate": 2.0230631749811306e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 36103768,
      "step": 22445
    },
    {
      "epoch": 1.0162973358382943,
      "grad_norm": 0.8034770488739014,
      "learning_rate": 2.0220994950086162e-05,
      "loss": 0.1536,
      "num_input_tokens_seen": 36111512,
      "step": 22450
    },
    {
      "epoch": 1.0165236877249373,
      "grad_norm": 0.6208401322364807,
      "learning_rate": 2.021135888734365e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 36119320,
      "step": 22455
    },
    {
      "epoch": 1.01675003961158,
      "grad_norm": 0.9259811043739319,
      "learning_rate": 2.0201723563069783e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 36128120,
      "step": 22460
    },
    {
      "epoch": 1.016976391498223,
      "grad_norm": 0.9097822904586792,
      "learning_rate": 2.0192088978750433e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 36136216,
      "step": 22465
    },
    {
      "epoch": 1.017202743384866,
      "grad_norm": 0.8164098858833313,
      "learning_rate": 2.0182455135871385e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 36143992,
      "step": 22470
    },
    {
      "epoch": 1.017429095271509,
      "grad_norm": 0.7025051712989807,
      "learning_rate": 2.0172822035918305e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 36152184,
      "step": 22475
    },
    {
      "epoch": 1.017655447158152,
      "grad_norm": 0.29398947954177856,
      "learning_rate": 2.016318968037671e-05,
      "loss": 0.1141,
      "num_input_tokens_seen": 36159864,
      "step": 22480
    },
    {
      "epoch": 1.017881799044795,
      "grad_norm": 1.7791017293930054,
      "learning_rate": 2.015355807073206e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 36167512,
      "step": 22485
    },
    {
      "epoch": 1.018108150931438,
      "grad_norm": 0.5523184537887573,
      "learning_rate": 2.0143927208469664e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 36175192,
      "step": 22490
    },
    {
      "epoch": 1.018334502818081,
      "grad_norm": 0.6161456108093262,
      "learning_rate": 2.0134297095074708e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 36183032,
      "step": 22495
    },
    {
      "epoch": 1.018560854704724,
      "grad_norm": 0.740833044052124,
      "learning_rate": 2.0124667732032297e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 36191576,
      "step": 22500
    },
    {
      "epoch": 1.018787206591367,
      "grad_norm": 0.5761888027191162,
      "learning_rate": 2.011503912082738e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 36199576,
      "step": 22505
    },
    {
      "epoch": 1.01901355847801,
      "grad_norm": 0.5626201629638672,
      "learning_rate": 2.0105411262944823e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 36207256,
      "step": 22510
    },
    {
      "epoch": 1.0192399103646528,
      "grad_norm": 0.49898719787597656,
      "learning_rate": 2.0095784159869366e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 36215512,
      "step": 22515
    },
    {
      "epoch": 1.0194662622512958,
      "grad_norm": 0.6718326807022095,
      "learning_rate": 2.0086157813085608e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 36223576,
      "step": 22520
    },
    {
      "epoch": 1.0196926141379388,
      "grad_norm": 0.4342135787010193,
      "learning_rate": 2.0076532224078068e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 36231576,
      "step": 22525
    },
    {
      "epoch": 1.0199189660245818,
      "grad_norm": 0.5164775848388672,
      "learning_rate": 2.0066907394331142e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 36239640,
      "step": 22530
    },
    {
      "epoch": 1.0201453179112248,
      "grad_norm": 0.5469797849655151,
      "learning_rate": 2.0057283325329077e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 36248120,
      "step": 22535
    },
    {
      "epoch": 1.0203716697978678,
      "grad_norm": 0.7618941068649292,
      "learning_rate": 2.0047660018556047e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 36255736,
      "step": 22540
    },
    {
      "epoch": 1.0205980216845107,
      "grad_norm": 0.6582110524177551,
      "learning_rate": 2.0038037475496075e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 36263640,
      "step": 22545
    },
    {
      "epoch": 1.0208243735711537,
      "grad_norm": 0.6284916996955872,
      "learning_rate": 2.0028415697633073e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 36271992,
      "step": 22550
    },
    {
      "epoch": 1.0210507254577967,
      "grad_norm": 0.88946932554245,
      "learning_rate": 2.0018794686450858e-05,
      "loss": 0.1935,
      "num_input_tokens_seen": 36279736,
      "step": 22555
    },
    {
      "epoch": 1.0212770773444397,
      "grad_norm": 0.46729063987731934,
      "learning_rate": 2.0009174443433088e-05,
      "loss": 0.187,
      "num_input_tokens_seen": 36287384,
      "step": 22560
    },
    {
      "epoch": 1.0215034292310827,
      "grad_norm": 0.5806694030761719,
      "learning_rate": 1.999955497006334e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 36296632,
      "step": 22565
    },
    {
      "epoch": 1.0217297811177257,
      "grad_norm": 0.827255129814148,
      "learning_rate": 1.9989936267825067e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 36304632,
      "step": 22570
    },
    {
      "epoch": 1.0219561330043685,
      "grad_norm": 0.6425843834877014,
      "learning_rate": 1.9980318338201572e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 36312888,
      "step": 22575
    },
    {
      "epoch": 1.0221824848910115,
      "grad_norm": 0.5107004642486572,
      "learning_rate": 1.997070118267607e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 36320344,
      "step": 22580
    },
    {
      "epoch": 1.0224088367776545,
      "grad_norm": 0.7176489233970642,
      "learning_rate": 1.9961084802731654e-05,
      "loss": 0.1155,
      "num_input_tokens_seen": 36327992,
      "step": 22585
    },
    {
      "epoch": 1.0226351886642975,
      "grad_norm": 0.8915581107139587,
      "learning_rate": 1.9951469199851273e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 36336120,
      "step": 22590
    },
    {
      "epoch": 1.0228615405509405,
      "grad_norm": 0.8355091214179993,
      "learning_rate": 1.99418543755178e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 36345464,
      "step": 22595
    },
    {
      "epoch": 1.0230878924375835,
      "grad_norm": 0.6601377129554749,
      "learning_rate": 1.9932240331213936e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 36353656,
      "step": 22600
    },
    {
      "epoch": 1.0230878924375835,
      "eval_loss": 0.14595334231853485,
      "eval_runtime": 404.4405,
      "eval_samples_per_second": 97.1,
      "eval_steps_per_second": 24.276,
      "num_input_tokens_seen": 36353656,
      "step": 22600
    },
    {
      "epoch": 1.0233142443242265,
      "grad_norm": 0.5244208574295044,
      "learning_rate": 1.9922627068422297e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 36361400,
      "step": 22605
    },
    {
      "epoch": 1.0235405962108695,
      "grad_norm": 0.6166813373565674,
      "learning_rate": 1.991301458862538e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 36369048,
      "step": 22610
    },
    {
      "epoch": 1.0237669480975125,
      "grad_norm": 0.39000648260116577,
      "learning_rate": 1.9903402893305536e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 36377016,
      "step": 22615
    },
    {
      "epoch": 1.0239932999841554,
      "grad_norm": 0.8733365535736084,
      "learning_rate": 1.9893791983945016e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 36385944,
      "step": 22620
    },
    {
      "epoch": 1.0242196518707984,
      "grad_norm": 0.4893893301486969,
      "learning_rate": 1.988418186202594e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 36393272,
      "step": 22625
    },
    {
      "epoch": 1.0244460037574412,
      "grad_norm": 0.556941568851471,
      "learning_rate": 1.98745725290303e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 36401560,
      "step": 22630
    },
    {
      "epoch": 1.0246723556440842,
      "grad_norm": 0.35039660334587097,
      "learning_rate": 1.986496398644e-05,
      "loss": 0.1449,
      "num_input_tokens_seen": 36408920,
      "step": 22635
    },
    {
      "epoch": 1.0248987075307272,
      "grad_norm": 0.418468713760376,
      "learning_rate": 1.9855356235736777e-05,
      "loss": 0.125,
      "num_input_tokens_seen": 36417496,
      "step": 22640
    },
    {
      "epoch": 1.0251250594173702,
      "grad_norm": 1.584753155708313,
      "learning_rate": 1.9845749278402277e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 36425432,
      "step": 22645
    },
    {
      "epoch": 1.0253514113040132,
      "grad_norm": 0.3722718060016632,
      "learning_rate": 1.9836143115918006e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 36432888,
      "step": 22650
    },
    {
      "epoch": 1.0255777631906562,
      "grad_norm": 1.124260663986206,
      "learning_rate": 1.9826537749765367e-05,
      "loss": 0.1287,
      "num_input_tokens_seen": 36440568,
      "step": 22655
    },
    {
      "epoch": 1.0258041150772992,
      "grad_norm": 0.4958493411540985,
      "learning_rate": 1.9816933181425625e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 36448696,
      "step": 22660
    },
    {
      "epoch": 1.0260304669639422,
      "grad_norm": 0.7468275427818298,
      "learning_rate": 1.9807329412379903e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 36456888,
      "step": 22665
    },
    {
      "epoch": 1.0262568188505852,
      "grad_norm": 1.0879478454589844,
      "learning_rate": 1.9797726444109247e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 36464760,
      "step": 22670
    },
    {
      "epoch": 1.0264831707372282,
      "grad_norm": 0.8426010012626648,
      "learning_rate": 1.9788124278094557e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 36472824,
      "step": 22675
    },
    {
      "epoch": 1.0267095226238712,
      "grad_norm": 0.46919068694114685,
      "learning_rate": 1.9778522915816594e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 36480408,
      "step": 22680
    },
    {
      "epoch": 1.0269358745105142,
      "grad_norm": 0.3743969798088074,
      "learning_rate": 1.9768922358756014e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 36488184,
      "step": 22685
    },
    {
      "epoch": 1.027162226397157,
      "grad_norm": 0.89670330286026,
      "learning_rate": 1.9759322608393353e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 36496056,
      "step": 22690
    },
    {
      "epoch": 1.0273885782838,
      "grad_norm": 0.6652807593345642,
      "learning_rate": 1.9749723666208992e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 36503896,
      "step": 22695
    },
    {
      "epoch": 1.027614930170443,
      "grad_norm": 1.5024102926254272,
      "learning_rate": 1.9740125533683235e-05,
      "loss": 0.164,
      "num_input_tokens_seen": 36512056,
      "step": 22700
    },
    {
      "epoch": 1.027841282057086,
      "grad_norm": 0.9512754082679749,
      "learning_rate": 1.9730528212296208e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 36520376,
      "step": 22705
    },
    {
      "epoch": 1.028067633943729,
      "grad_norm": 0.34029412269592285,
      "learning_rate": 1.9720931703527945e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 36528952,
      "step": 22710
    },
    {
      "epoch": 1.028293985830372,
      "grad_norm": 0.4215284585952759,
      "learning_rate": 1.9711336008858373e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 36536568,
      "step": 22715
    },
    {
      "epoch": 1.028520337717015,
      "grad_norm": 1.2123850584030151,
      "learning_rate": 1.9701741129767233e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 36545080,
      "step": 22720
    },
    {
      "epoch": 1.028746689603658,
      "grad_norm": 0.45927998423576355,
      "learning_rate": 1.9692147067734202e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 36553624,
      "step": 22725
    },
    {
      "epoch": 1.0289730414903009,
      "grad_norm": 0.5517261028289795,
      "learning_rate": 1.96825538242388e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 36561720,
      "step": 22730
    },
    {
      "epoch": 1.0291993933769439,
      "grad_norm": 0.38795191049575806,
      "learning_rate": 1.967296140076041e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 36569208,
      "step": 22735
    },
    {
      "epoch": 1.0294257452635869,
      "grad_norm": 0.7346002459526062,
      "learning_rate": 1.966336979877833e-05,
      "loss": 0.1228,
      "num_input_tokens_seen": 36577304,
      "step": 22740
    },
    {
      "epoch": 1.0296520971502297,
      "grad_norm": 0.5559272170066833,
      "learning_rate": 1.9653779019771678e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 36585304,
      "step": 22745
    },
    {
      "epoch": 1.0298784490368726,
      "grad_norm": 0.6687783002853394,
      "learning_rate": 1.9644189065219488e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 36592472,
      "step": 22750
    },
    {
      "epoch": 1.0301048009235156,
      "grad_norm": 0.7287868857383728,
      "learning_rate": 1.9634599936600655e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 36600088,
      "step": 22755
    },
    {
      "epoch": 1.0303311528101586,
      "grad_norm": 0.37214094400405884,
      "learning_rate": 1.9625011635393935e-05,
      "loss": 0.1204,
      "num_input_tokens_seen": 36607864,
      "step": 22760
    },
    {
      "epoch": 1.0305575046968016,
      "grad_norm": 0.840452253818512,
      "learning_rate": 1.9615424163077963e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 36615416,
      "step": 22765
    },
    {
      "epoch": 1.0307838565834446,
      "grad_norm": 0.799527645111084,
      "learning_rate": 1.9605837521131263e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 36623576,
      "step": 22770
    },
    {
      "epoch": 1.0310102084700876,
      "grad_norm": 0.4578310251235962,
      "learning_rate": 1.9596251711032192e-05,
      "loss": 0.1557,
      "num_input_tokens_seen": 36632120,
      "step": 22775
    },
    {
      "epoch": 1.0312365603567306,
      "grad_norm": 1.2585989236831665,
      "learning_rate": 1.958666673425903e-05,
      "loss": 0.2122,
      "num_input_tokens_seen": 36640216,
      "step": 22780
    },
    {
      "epoch": 1.0314629122433736,
      "grad_norm": 0.7041440010070801,
      "learning_rate": 1.957708259228987e-05,
      "loss": 0.1583,
      "num_input_tokens_seen": 36648120,
      "step": 22785
    },
    {
      "epoch": 1.0316892641300166,
      "grad_norm": 0.6396097540855408,
      "learning_rate": 1.956749928660273e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 36656312,
      "step": 22790
    },
    {
      "epoch": 1.0319156160166596,
      "grad_norm": 0.5892245173454285,
      "learning_rate": 1.955791681867547e-05,
      "loss": 0.1716,
      "num_input_tokens_seen": 36664024,
      "step": 22795
    },
    {
      "epoch": 1.0321419679033024,
      "grad_norm": 0.7498739361763,
      "learning_rate": 1.9548335189985824e-05,
      "loss": 0.157,
      "num_input_tokens_seen": 36671480,
      "step": 22800
    },
    {
      "epoch": 1.0321419679033024,
      "eval_loss": 0.14570175111293793,
      "eval_runtime": 404.2588,
      "eval_samples_per_second": 97.143,
      "eval_steps_per_second": 24.286,
      "num_input_tokens_seen": 36671480,
      "step": 22800
    },
    {
      "epoch": 1.0323683197899454,
      "grad_norm": 0.5091695785522461,
      "learning_rate": 1.9538754402011396e-05,
      "loss": 0.1669,
      "num_input_tokens_seen": 36679928,
      "step": 22805
    },
    {
      "epoch": 1.0325946716765884,
      "grad_norm": 0.9657270908355713,
      "learning_rate": 1.952917445622968e-05,
      "loss": 0.1963,
      "num_input_tokens_seen": 36687608,
      "step": 22810
    },
    {
      "epoch": 1.0328210235632314,
      "grad_norm": 0.559059739112854,
      "learning_rate": 1.9519595354118005e-05,
      "loss": 0.119,
      "num_input_tokens_seen": 36695320,
      "step": 22815
    },
    {
      "epoch": 1.0330473754498743,
      "grad_norm": 0.707736611366272,
      "learning_rate": 1.951001709715361e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 36703832,
      "step": 22820
    },
    {
      "epoch": 1.0332737273365173,
      "grad_norm": 0.9487020373344421,
      "learning_rate": 1.9500439686813556e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 36711800,
      "step": 22825
    },
    {
      "epoch": 1.0335000792231603,
      "grad_norm": 0.521578311920166,
      "learning_rate": 1.949086312457482e-05,
      "loss": 0.1732,
      "num_input_tokens_seen": 36719352,
      "step": 22830
    },
    {
      "epoch": 1.0337264311098033,
      "grad_norm": 0.5791155099868774,
      "learning_rate": 1.9481287411914223e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 36727640,
      "step": 22835
    },
    {
      "epoch": 1.0339527829964463,
      "grad_norm": 0.9017087817192078,
      "learning_rate": 1.9471712550308457e-05,
      "loss": 0.1348,
      "num_input_tokens_seen": 36735960,
      "step": 22840
    },
    {
      "epoch": 1.0341791348830893,
      "grad_norm": 0.27871692180633545,
      "learning_rate": 1.946213854123409e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 36743576,
      "step": 22845
    },
    {
      "epoch": 1.0344054867697323,
      "grad_norm": 0.6268014907836914,
      "learning_rate": 1.9452565386167554e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 36751448,
      "step": 22850
    },
    {
      "epoch": 1.034631838656375,
      "grad_norm": 0.3852930963039398,
      "learning_rate": 1.9442993086585142e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 36758968,
      "step": 22855
    },
    {
      "epoch": 1.034858190543018,
      "grad_norm": 0.44129425287246704,
      "learning_rate": 1.9433421643963043e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 36767256,
      "step": 22860
    },
    {
      "epoch": 1.035084542429661,
      "grad_norm": 1.3014607429504395,
      "learning_rate": 1.942385105977727e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 36774776,
      "step": 22865
    },
    {
      "epoch": 1.035310894316304,
      "grad_norm": 0.5528838038444519,
      "learning_rate": 1.9414281335503743e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 36782296,
      "step": 22870
    },
    {
      "epoch": 1.035537246202947,
      "grad_norm": 0.5166964530944824,
      "learning_rate": 1.9404712472618232e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 36790008,
      "step": 22875
    },
    {
      "epoch": 1.03576359808959,
      "grad_norm": 0.7595129013061523,
      "learning_rate": 1.939514447259636e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 36798296,
      "step": 22880
    },
    {
      "epoch": 1.035989949976233,
      "grad_norm": 0.9870322346687317,
      "learning_rate": 1.938557733691365e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 36806456,
      "step": 22885
    },
    {
      "epoch": 1.036216301862876,
      "grad_norm": 0.6038070917129517,
      "learning_rate": 1.9376011067045476e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 36814168,
      "step": 22890
    },
    {
      "epoch": 1.036442653749519,
      "grad_norm": 0.8797662258148193,
      "learning_rate": 1.9366445664467065e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 36822232,
      "step": 22895
    },
    {
      "epoch": 1.036669005636162,
      "grad_norm": 0.3207445442676544,
      "learning_rate": 1.9356881130653533e-05,
      "loss": 0.1164,
      "num_input_tokens_seen": 36830616,
      "step": 22900
    },
    {
      "epoch": 1.036895357522805,
      "grad_norm": 0.33830276131629944,
      "learning_rate": 1.9347317467079846e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 36838264,
      "step": 22905
    },
    {
      "epoch": 1.037121709409448,
      "grad_norm": 0.5820893049240112,
      "learning_rate": 1.9337754675220836e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 36846232,
      "step": 22910
    },
    {
      "epoch": 1.0373480612960908,
      "grad_norm": 0.31889694929122925,
      "learning_rate": 1.9328192756551218e-05,
      "loss": 0.1201,
      "num_input_tokens_seen": 36854200,
      "step": 22915
    },
    {
      "epoch": 1.0375744131827338,
      "grad_norm": 0.6232013702392578,
      "learning_rate": 1.931863171254555e-05,
      "loss": 0.1106,
      "num_input_tokens_seen": 36862584,
      "step": 22920
    },
    {
      "epoch": 1.0378007650693768,
      "grad_norm": 0.586234450340271,
      "learning_rate": 1.930907154467826e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 36871256,
      "step": 22925
    },
    {
      "epoch": 1.0380271169560198,
      "grad_norm": 0.9838700294494629,
      "learning_rate": 1.9299512254423673e-05,
      "loss": 0.1374,
      "num_input_tokens_seen": 36880088,
      "step": 22930
    },
    {
      "epoch": 1.0382534688426628,
      "grad_norm": 0.5644727349281311,
      "learning_rate": 1.9289953843255914e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 36887608,
      "step": 22935
    },
    {
      "epoch": 1.0384798207293058,
      "grad_norm": 0.6258010268211365,
      "learning_rate": 1.9280396312649048e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 36895416,
      "step": 22940
    },
    {
      "epoch": 1.0387061726159488,
      "grad_norm": 0.676202118396759,
      "learning_rate": 1.9270839664076936e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 36903800,
      "step": 22945
    },
    {
      "epoch": 1.0389325245025918,
      "grad_norm": 0.7432723045349121,
      "learning_rate": 1.9261283899013345e-05,
      "loss": 0.1238,
      "num_input_tokens_seen": 36911320,
      "step": 22950
    },
    {
      "epoch": 1.0391588763892348,
      "grad_norm": 0.4982748031616211,
      "learning_rate": 1.92517290189319e-05,
      "loss": 0.1118,
      "num_input_tokens_seen": 36918968,
      "step": 22955
    },
    {
      "epoch": 1.0393852282758778,
      "grad_norm": 0.6497018933296204,
      "learning_rate": 1.924217502530607e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 36927608,
      "step": 22960
    },
    {
      "epoch": 1.0396115801625208,
      "grad_norm": 1.1911842823028564,
      "learning_rate": 1.9232621919609207e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 36935320,
      "step": 22965
    },
    {
      "epoch": 1.0398379320491635,
      "grad_norm": 0.5705532431602478,
      "learning_rate": 1.9223069703314534e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 36943256,
      "step": 22970
    },
    {
      "epoch": 1.0400642839358065,
      "grad_norm": 0.517217218875885,
      "learning_rate": 1.92135183778951e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 36951064,
      "step": 22975
    },
    {
      "epoch": 1.0402906358224495,
      "grad_norm": 0.8901706337928772,
      "learning_rate": 1.9203967944823857e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 36958744,
      "step": 22980
    },
    {
      "epoch": 1.0405169877090925,
      "grad_norm": 0.9117767810821533,
      "learning_rate": 1.9194418405573588e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 36966712,
      "step": 22985
    },
    {
      "epoch": 1.0407433395957355,
      "grad_norm": 0.3777430057525635,
      "learning_rate": 1.9184869761616954e-05,
      "loss": 0.158,
      "num_input_tokens_seen": 36974712,
      "step": 22990
    },
    {
      "epoch": 1.0409696914823785,
      "grad_norm": 0.7545064687728882,
      "learning_rate": 1.9175322014426495e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 36982776,
      "step": 22995
    },
    {
      "epoch": 1.0411960433690215,
      "grad_norm": 1.026389718055725,
      "learning_rate": 1.9165775165474565e-05,
      "loss": 0.1168,
      "num_input_tokens_seen": 36990680,
      "step": 23000
    },
    {
      "epoch": 1.0411960433690215,
      "eval_loss": 0.14655929803848267,
      "eval_runtime": 404.407,
      "eval_samples_per_second": 97.108,
      "eval_steps_per_second": 24.278,
      "num_input_tokens_seen": 36990680,
      "step": 23000
    },
    {
      "epoch": 1.0414223952556645,
      "grad_norm": 1.1254949569702148,
      "learning_rate": 1.9156229216233434e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 36998040,
      "step": 23005
    },
    {
      "epoch": 1.0416487471423075,
      "grad_norm": 1.1286643743515015,
      "learning_rate": 1.9146684168175184e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 37006744,
      "step": 23010
    },
    {
      "epoch": 1.0418750990289505,
      "grad_norm": 0.6972285509109497,
      "learning_rate": 1.9137140022771796e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 37014648,
      "step": 23015
    },
    {
      "epoch": 1.0421014509155935,
      "grad_norm": 0.48021262884140015,
      "learning_rate": 1.9127596781495103e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 37022360,
      "step": 23020
    },
    {
      "epoch": 1.0423278028022365,
      "grad_norm": 0.817145586013794,
      "learning_rate": 1.9118054445816767e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 37031256,
      "step": 23025
    },
    {
      "epoch": 1.0425541546888792,
      "grad_norm": 0.5728305578231812,
      "learning_rate": 1.9108513017208356e-05,
      "loss": 0.1593,
      "num_input_tokens_seen": 37039352,
      "step": 23030
    },
    {
      "epoch": 1.0427805065755222,
      "grad_norm": 0.4167305529117584,
      "learning_rate": 1.9098972497141287e-05,
      "loss": 0.182,
      "num_input_tokens_seen": 37046904,
      "step": 23035
    },
    {
      "epoch": 1.0430068584621652,
      "grad_norm": 1.1329635381698608,
      "learning_rate": 1.9089432887086806e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 37055064,
      "step": 23040
    },
    {
      "epoch": 1.0432332103488082,
      "grad_norm": 0.3950503170490265,
      "learning_rate": 1.9079894188516056e-05,
      "loss": 0.1127,
      "num_input_tokens_seen": 37062680,
      "step": 23045
    },
    {
      "epoch": 1.0434595622354512,
      "grad_norm": 1.0586271286010742,
      "learning_rate": 1.907035640290002e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 37070520,
      "step": 23050
    },
    {
      "epoch": 1.0436859141220942,
      "grad_norm": 0.7628342509269714,
      "learning_rate": 1.9060819531709534e-05,
      "loss": 0.1247,
      "num_input_tokens_seen": 37078872,
      "step": 23055
    },
    {
      "epoch": 1.0439122660087372,
      "grad_norm": 0.8718791604042053,
      "learning_rate": 1.9051283576415325e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 37086680,
      "step": 23060
    },
    {
      "epoch": 1.0441386178953802,
      "grad_norm": 0.49335092306137085,
      "learning_rate": 1.904174853848793e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 37094648,
      "step": 23065
    },
    {
      "epoch": 1.0443649697820232,
      "grad_norm": 0.8665481209754944,
      "learning_rate": 1.903221441939779e-05,
      "loss": 0.2005,
      "num_input_tokens_seen": 37102360,
      "step": 23070
    },
    {
      "epoch": 1.0445913216686662,
      "grad_norm": 0.3666498064994812,
      "learning_rate": 1.9022681220615194e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 37109976,
      "step": 23075
    },
    {
      "epoch": 1.0448176735553092,
      "grad_norm": 1.3456720113754272,
      "learning_rate": 1.9013148943610255e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 37118232,
      "step": 23080
    },
    {
      "epoch": 1.045044025441952,
      "grad_norm": 0.6549904346466064,
      "learning_rate": 1.9003617589852998e-05,
      "loss": 0.1616,
      "num_input_tokens_seen": 37126232,
      "step": 23085
    },
    {
      "epoch": 1.045270377328595,
      "grad_norm": 0.404493123292923,
      "learning_rate": 1.899408716081326e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 37134360,
      "step": 23090
    },
    {
      "epoch": 1.045496729215238,
      "grad_norm": 0.34186094999313354,
      "learning_rate": 1.898455765796075e-05,
      "loss": 0.1396,
      "num_input_tokens_seen": 37142680,
      "step": 23095
    },
    {
      "epoch": 1.045723081101881,
      "grad_norm": 0.9118211269378662,
      "learning_rate": 1.8975029082765053e-05,
      "loss": 0.1596,
      "num_input_tokens_seen": 37150520,
      "step": 23100
    },
    {
      "epoch": 1.045949432988524,
      "grad_norm": 0.35049304366111755,
      "learning_rate": 1.8965501436695577e-05,
      "loss": 0.1327,
      "num_input_tokens_seen": 37157624,
      "step": 23105
    },
    {
      "epoch": 1.046175784875167,
      "grad_norm": 0.6058930158615112,
      "learning_rate": 1.895597472122161e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 37166104,
      "step": 23110
    },
    {
      "epoch": 1.04640213676181,
      "grad_norm": 0.363856703042984,
      "learning_rate": 1.894644893781231e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 37173560,
      "step": 23115
    },
    {
      "epoch": 1.046628488648453,
      "grad_norm": 0.22641852498054504,
      "learning_rate": 1.893692408793665e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 37181208,
      "step": 23120
    },
    {
      "epoch": 1.046854840535096,
      "grad_norm": 0.3792927861213684,
      "learning_rate": 1.8927400173063493e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 37189304,
      "step": 23125
    },
    {
      "epoch": 1.047081192421739,
      "grad_norm": 0.4081237316131592,
      "learning_rate": 1.891787719466154e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 37196920,
      "step": 23130
    },
    {
      "epoch": 1.047307544308382,
      "grad_norm": 0.5220649838447571,
      "learning_rate": 1.8908355154199346e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 37205464,
      "step": 23135
    },
    {
      "epoch": 1.0475338961950247,
      "grad_norm": 0.4567030370235443,
      "learning_rate": 1.8898834053145357e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 37214424,
      "step": 23140
    },
    {
      "epoch": 1.0477602480816677,
      "grad_norm": 0.5531008839607239,
      "learning_rate": 1.8889313892967813e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 37222488,
      "step": 23145
    },
    {
      "epoch": 1.0479865999683107,
      "grad_norm": 0.314720094203949,
      "learning_rate": 1.8879794675134863e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 37230008,
      "step": 23150
    },
    {
      "epoch": 1.0482129518549537,
      "grad_norm": 0.9756635427474976,
      "learning_rate": 1.8870276401114494e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 37238392,
      "step": 23155
    },
    {
      "epoch": 1.0484393037415967,
      "grad_norm": 0.583401083946228,
      "learning_rate": 1.886075907237453e-05,
      "loss": 0.1157,
      "num_input_tokens_seen": 37246904,
      "step": 23160
    },
    {
      "epoch": 1.0486656556282397,
      "grad_norm": 0.6911195516586304,
      "learning_rate": 1.8851242690382672e-05,
      "loss": 0.1713,
      "num_input_tokens_seen": 37255352,
      "step": 23165
    },
    {
      "epoch": 1.0488920075148827,
      "grad_norm": 1.0943809747695923,
      "learning_rate": 1.884172725660645e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 37263576,
      "step": 23170
    },
    {
      "epoch": 1.0491183594015256,
      "grad_norm": 0.5900644659996033,
      "learning_rate": 1.8832212772513277e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 37271736,
      "step": 23175
    },
    {
      "epoch": 1.0493447112881686,
      "grad_norm": 0.4906558096408844,
      "learning_rate": 1.8822699239570414e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 37279832,
      "step": 23180
    },
    {
      "epoch": 1.0495710631748116,
      "grad_norm": 1.280059576034546,
      "learning_rate": 1.8813186659244943e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 37288088,
      "step": 23185
    },
    {
      "epoch": 1.0497974150614546,
      "grad_norm": 1.0700645446777344,
      "learning_rate": 1.880367503300385e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 37295992,
      "step": 23190
    },
    {
      "epoch": 1.0500237669480974,
      "grad_norm": 0.7171104550361633,
      "learning_rate": 1.8794164362313927e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 37304312,
      "step": 23195
    },
    {
      "epoch": 1.0502501188347404,
      "grad_norm": 0.6491122841835022,
      "learning_rate": 1.878465464864185e-05,
      "loss": 0.0976,
      "num_input_tokens_seen": 37312248,
      "step": 23200
    },
    {
      "epoch": 1.0502501188347404,
      "eval_loss": 0.145964115858078,
      "eval_runtime": 404.5056,
      "eval_samples_per_second": 97.084,
      "eval_steps_per_second": 24.272,
      "num_input_tokens_seen": 37312248,
      "step": 23200
    },
    {
      "epoch": 1.0504764707213834,
      "grad_norm": 0.7575937509536743,
      "learning_rate": 1.877514589345414e-05,
      "loss": 0.1171,
      "num_input_tokens_seen": 37320504,
      "step": 23205
    },
    {
      "epoch": 1.0507028226080264,
      "grad_norm": 0.4312746524810791,
      "learning_rate": 1.876563809821715e-05,
      "loss": 0.1036,
      "num_input_tokens_seen": 37328216,
      "step": 23210
    },
    {
      "epoch": 1.0509291744946694,
      "grad_norm": 1.0092335939407349,
      "learning_rate": 1.8756131264397106e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 37336600,
      "step": 23215
    },
    {
      "epoch": 1.0511555263813124,
      "grad_norm": 0.39715245366096497,
      "learning_rate": 1.87466253934601e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 37344376,
      "step": 23220
    },
    {
      "epoch": 1.0513818782679554,
      "grad_norm": 0.3861795961856842,
      "learning_rate": 1.8737120486872033e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 37352696,
      "step": 23225
    },
    {
      "epoch": 1.0516082301545984,
      "grad_norm": 0.8365803360939026,
      "learning_rate": 1.8727616546098696e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 37361464,
      "step": 23230
    },
    {
      "epoch": 1.0518345820412414,
      "grad_norm": 0.4310343563556671,
      "learning_rate": 1.8718113572605716e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 37369048,
      "step": 23235
    },
    {
      "epoch": 1.0520609339278844,
      "grad_norm": 0.581218957901001,
      "learning_rate": 1.8708611567858554e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 37377336,
      "step": 23240
    },
    {
      "epoch": 1.0522872858145274,
      "grad_norm": 0.9425561428070068,
      "learning_rate": 1.8699110533322565e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 37385816,
      "step": 23245
    },
    {
      "epoch": 1.0525136377011703,
      "grad_norm": 0.4891781806945801,
      "learning_rate": 1.8689610470462897e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 37393912,
      "step": 23250
    },
    {
      "epoch": 1.0527399895878131,
      "grad_norm": 0.47282904386520386,
      "learning_rate": 1.8680111380744604e-05,
      "loss": 0.1264,
      "num_input_tokens_seen": 37401784,
      "step": 23255
    },
    {
      "epoch": 1.0529663414744561,
      "grad_norm": 0.6079865097999573,
      "learning_rate": 1.8670613265632564e-05,
      "loss": 0.1261,
      "num_input_tokens_seen": 37408856,
      "step": 23260
    },
    {
      "epoch": 1.053192693361099,
      "grad_norm": 0.8306401371955872,
      "learning_rate": 1.866111612659149e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 37417304,
      "step": 23265
    },
    {
      "epoch": 1.053419045247742,
      "grad_norm": 0.5538521409034729,
      "learning_rate": 1.8651619965085967e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 37425880,
      "step": 23270
    },
    {
      "epoch": 1.053645397134385,
      "grad_norm": 0.8916104435920715,
      "learning_rate": 1.8642124782580433e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 37434040,
      "step": 23275
    },
    {
      "epoch": 1.053871749021028,
      "grad_norm": 0.6665787696838379,
      "learning_rate": 1.8632630580539144e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 37441912,
      "step": 23280
    },
    {
      "epoch": 1.054098100907671,
      "grad_norm": 0.591568112373352,
      "learning_rate": 1.862313736042625e-05,
      "loss": 0.1424,
      "num_input_tokens_seen": 37450328,
      "step": 23285
    },
    {
      "epoch": 1.054324452794314,
      "grad_norm": 0.47428232431411743,
      "learning_rate": 1.8613645123705703e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 37458424,
      "step": 23290
    },
    {
      "epoch": 1.054550804680957,
      "grad_norm": 0.6403087377548218,
      "learning_rate": 1.8604153871841328e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 37466200,
      "step": 23295
    },
    {
      "epoch": 1.0547771565676,
      "grad_norm": 0.5524865388870239,
      "learning_rate": 1.859466360629682e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 37474136,
      "step": 23300
    },
    {
      "epoch": 1.055003508454243,
      "grad_norm": 0.7498432397842407,
      "learning_rate": 1.8585174328535666e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 37481784,
      "step": 23305
    },
    {
      "epoch": 1.0552298603408858,
      "grad_norm": 0.7941622734069824,
      "learning_rate": 1.857568604002124e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 37489112,
      "step": 23310
    },
    {
      "epoch": 1.0554562122275288,
      "grad_norm": 0.6651421189308167,
      "learning_rate": 1.8566198742216774e-05,
      "loss": 0.135,
      "num_input_tokens_seen": 37497080,
      "step": 23315
    },
    {
      "epoch": 1.0556825641141718,
      "grad_norm": 0.4325953722000122,
      "learning_rate": 1.85567124365853e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 37505272,
      "step": 23320
    },
    {
      "epoch": 1.0559089160008148,
      "grad_norm": 0.5009159445762634,
      "learning_rate": 1.854722712458975e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 37512856,
      "step": 23325
    },
    {
      "epoch": 1.0561352678874578,
      "grad_norm": 0.6905559301376343,
      "learning_rate": 1.853774280769286e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 37520568,
      "step": 23330
    },
    {
      "epoch": 1.0563616197741008,
      "grad_norm": 0.8436486721038818,
      "learning_rate": 1.852825948735724e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 37528248,
      "step": 23335
    },
    {
      "epoch": 1.0565879716607438,
      "grad_norm": 1.01020348072052,
      "learning_rate": 1.851877716504534e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 37536120,
      "step": 23340
    },
    {
      "epoch": 1.0568143235473868,
      "grad_norm": 0.7261248826980591,
      "learning_rate": 1.8509295842219448e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 37543736,
      "step": 23345
    },
    {
      "epoch": 1.0570406754340298,
      "grad_norm": 0.5777470469474792,
      "learning_rate": 1.8499815520341697e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 37550936,
      "step": 23350
    },
    {
      "epoch": 1.0572670273206728,
      "grad_norm": 0.6808956265449524,
      "learning_rate": 1.8490336200874094e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 37559096,
      "step": 23355
    },
    {
      "epoch": 1.0574933792073158,
      "grad_norm": 0.35407835245132446,
      "learning_rate": 1.848085788527844e-05,
      "loss": 0.1115,
      "num_input_tokens_seen": 37567576,
      "step": 23360
    },
    {
      "epoch": 1.0577197310939588,
      "grad_norm": 0.44435879588127136,
      "learning_rate": 1.847138057501644e-05,
      "loss": 0.1137,
      "num_input_tokens_seen": 37575864,
      "step": 23365
    },
    {
      "epoch": 1.0579460829806016,
      "grad_norm": 0.4829294979572296,
      "learning_rate": 1.8461904271549582e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 37584376,
      "step": 23370
    },
    {
      "epoch": 1.0581724348672445,
      "grad_norm": 0.6985398530960083,
      "learning_rate": 1.845242897633926e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 37592312,
      "step": 23375
    },
    {
      "epoch": 1.0583987867538875,
      "grad_norm": 0.49010515213012695,
      "learning_rate": 1.844295469084667e-05,
      "loss": 0.1694,
      "num_input_tokens_seen": 37600344,
      "step": 23380
    },
    {
      "epoch": 1.0586251386405305,
      "grad_norm": 0.37377244234085083,
      "learning_rate": 1.843348141653286e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 37608216,
      "step": 23385
    },
    {
      "epoch": 1.0588514905271735,
      "grad_norm": 0.9005696773529053,
      "learning_rate": 1.842400915485874e-05,
      "loss": 0.1575,
      "num_input_tokens_seen": 37616792,
      "step": 23390
    },
    {
      "epoch": 1.0590778424138165,
      "grad_norm": 0.5274709463119507,
      "learning_rate": 1.8414537907285053e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 37625432,
      "step": 23395
    },
    {
      "epoch": 1.0593041943004595,
      "grad_norm": 0.3862432837486267,
      "learning_rate": 1.840506767527237e-05,
      "loss": 0.1567,
      "num_input_tokens_seen": 37633848,
      "step": 23400
    },
    {
      "epoch": 1.0593041943004595,
      "eval_loss": 0.14574775099754333,
      "eval_runtime": 404.8511,
      "eval_samples_per_second": 97.001,
      "eval_steps_per_second": 24.251,
      "num_input_tokens_seen": 37633848,
      "step": 23400
    },
    {
      "epoch": 1.0595305461871025,
      "grad_norm": 0.6609634160995483,
      "learning_rate": 1.8395598460281137e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 37642008,
      "step": 23405
    },
    {
      "epoch": 1.0597568980737455,
      "grad_norm": 0.4713321030139923,
      "learning_rate": 1.838613026377161e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 37650200,
      "step": 23410
    },
    {
      "epoch": 1.0599832499603885,
      "grad_norm": 1.1374573707580566,
      "learning_rate": 1.8376663087203917e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 37657656,
      "step": 23415
    },
    {
      "epoch": 1.0602096018470315,
      "grad_norm": 1.7184184789657593,
      "learning_rate": 1.8367196932038014e-05,
      "loss": 0.1676,
      "num_input_tokens_seen": 37665912,
      "step": 23420
    },
    {
      "epoch": 1.0604359537336743,
      "grad_norm": 0.8121074438095093,
      "learning_rate": 1.8357731799733686e-05,
      "loss": 0.1795,
      "num_input_tokens_seen": 37673656,
      "step": 23425
    },
    {
      "epoch": 1.0606623056203173,
      "grad_norm": 0.804559051990509,
      "learning_rate": 1.8348267691750586e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 37682104,
      "step": 23430
    },
    {
      "epoch": 1.0608886575069603,
      "grad_norm": 0.6830423474311829,
      "learning_rate": 1.833880460954821e-05,
      "loss": 0.1231,
      "num_input_tokens_seen": 37689624,
      "step": 23435
    },
    {
      "epoch": 1.0611150093936033,
      "grad_norm": 0.3745306432247162,
      "learning_rate": 1.8329342554585866e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 37697368,
      "step": 23440
    },
    {
      "epoch": 1.0613413612802463,
      "grad_norm": 0.9793573617935181,
      "learning_rate": 1.8319881528322735e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 37705336,
      "step": 23445
    },
    {
      "epoch": 1.0615677131668892,
      "grad_norm": 0.7562173008918762,
      "learning_rate": 1.8310421532217815e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 37713048,
      "step": 23450
    },
    {
      "epoch": 1.0617940650535322,
      "grad_norm": 0.9781950116157532,
      "learning_rate": 1.8300962567729958e-05,
      "loss": 0.179,
      "num_input_tokens_seen": 37721208,
      "step": 23455
    },
    {
      "epoch": 1.0620204169401752,
      "grad_norm": 0.3278777301311493,
      "learning_rate": 1.8291504636317866e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 37729112,
      "step": 23460
    },
    {
      "epoch": 1.0622467688268182,
      "grad_norm": 1.0700815916061401,
      "learning_rate": 1.8282047739440055e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 37737016,
      "step": 23465
    },
    {
      "epoch": 1.0624731207134612,
      "grad_norm": 0.49445655941963196,
      "learning_rate": 1.8272591878554903e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 37745176,
      "step": 23470
    },
    {
      "epoch": 1.0626994726001042,
      "grad_norm": 0.7581466436386108,
      "learning_rate": 1.8263137055120638e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 37753784,
      "step": 23475
    },
    {
      "epoch": 1.062925824486747,
      "grad_norm": 0.7074391841888428,
      "learning_rate": 1.8253683270595295e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 37761816,
      "step": 23480
    },
    {
      "epoch": 1.06315217637339,
      "grad_norm": 0.39582639932632446,
      "learning_rate": 1.824423052643677e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 37769848,
      "step": 23485
    },
    {
      "epoch": 1.063378528260033,
      "grad_norm": 0.6320398449897766,
      "learning_rate": 1.82347788241028e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 37778072,
      "step": 23490
    },
    {
      "epoch": 1.063604880146676,
      "grad_norm": 0.8079730868339539,
      "learning_rate": 1.8225328165050942e-05,
      "loss": 0.1753,
      "num_input_tokens_seen": 37785592,
      "step": 23495
    },
    {
      "epoch": 1.063831232033319,
      "grad_norm": 0.4544483423233032,
      "learning_rate": 1.821587855073863e-05,
      "loss": 0.189,
      "num_input_tokens_seen": 37794200,
      "step": 23500
    },
    {
      "epoch": 1.064057583919962,
      "grad_norm": 0.8241644501686096,
      "learning_rate": 1.8206429982623086e-05,
      "loss": 0.1336,
      "num_input_tokens_seen": 37802456,
      "step": 23505
    },
    {
      "epoch": 1.064283935806605,
      "grad_norm": 1.0841665267944336,
      "learning_rate": 1.8196982462161416e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 37810360,
      "step": 23510
    },
    {
      "epoch": 1.064510287693248,
      "grad_norm": 0.3677912950515747,
      "learning_rate": 1.818753599081055e-05,
      "loss": 0.1206,
      "num_input_tokens_seen": 37818136,
      "step": 23515
    },
    {
      "epoch": 1.064736639579891,
      "grad_norm": 0.4677456319332123,
      "learning_rate": 1.817809057002724e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 37825944,
      "step": 23520
    },
    {
      "epoch": 1.064962991466534,
      "grad_norm": 0.9698733687400818,
      "learning_rate": 1.8168646201268096e-05,
      "loss": 0.2186,
      "num_input_tokens_seen": 37834104,
      "step": 23525
    },
    {
      "epoch": 1.065189343353177,
      "grad_norm": 0.464489609003067,
      "learning_rate": 1.8159202885989557e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 37841592,
      "step": 23530
    },
    {
      "epoch": 1.0654156952398197,
      "grad_norm": 0.8727866411209106,
      "learning_rate": 1.814976062564789e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 37849688,
      "step": 23535
    },
    {
      "epoch": 1.0656420471264627,
      "grad_norm": 0.477840781211853,
      "learning_rate": 1.8140319421699234e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 37857784,
      "step": 23540
    },
    {
      "epoch": 1.0658683990131057,
      "grad_norm": 0.4623295068740845,
      "learning_rate": 1.8130879275599515e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 37866072,
      "step": 23545
    },
    {
      "epoch": 1.0660947508997487,
      "grad_norm": 0.5304109454154968,
      "learning_rate": 1.8121440188804544e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 37873944,
      "step": 23550
    },
    {
      "epoch": 1.0663211027863917,
      "grad_norm": 0.5275995135307312,
      "learning_rate": 1.811200216276993e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 37881976,
      "step": 23555
    },
    {
      "epoch": 1.0665474546730347,
      "grad_norm": 0.4262065291404724,
      "learning_rate": 1.810256519895115e-05,
      "loss": 0.1576,
      "num_input_tokens_seen": 37890872,
      "step": 23560
    },
    {
      "epoch": 1.0667738065596777,
      "grad_norm": 0.1972026228904724,
      "learning_rate": 1.8093129298803494e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 37898936,
      "step": 23565
    },
    {
      "epoch": 1.0670001584463207,
      "grad_norm": 0.36357298493385315,
      "learning_rate": 1.808369446378209e-05,
      "loss": 0.148,
      "num_input_tokens_seen": 37906360,
      "step": 23570
    },
    {
      "epoch": 1.0672265103329637,
      "grad_norm": 0.4732073247432709,
      "learning_rate": 1.8074260695341914e-05,
      "loss": 0.1602,
      "num_input_tokens_seen": 37914008,
      "step": 23575
    },
    {
      "epoch": 1.0674528622196067,
      "grad_norm": 0.31963297724723816,
      "learning_rate": 1.8064827994937782e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 37922392,
      "step": 23580
    },
    {
      "epoch": 1.0676792141062497,
      "grad_norm": 0.9999583959579468,
      "learning_rate": 1.8055396364024317e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 37930328,
      "step": 23585
    },
    {
      "epoch": 1.0679055659928927,
      "grad_norm": 0.6830312609672546,
      "learning_rate": 1.804596580405601e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 37938744,
      "step": 23590
    },
    {
      "epoch": 1.0681319178795354,
      "grad_norm": 0.9783816933631897,
      "learning_rate": 1.8036536316487174e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 37946232,
      "step": 23595
    },
    {
      "epoch": 1.0683582697661784,
      "grad_norm": 0.593778133392334,
      "learning_rate": 1.802710790277193e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 37954104,
      "step": 23600
    },
    {
      "epoch": 1.0683582697661784,
      "eval_loss": 0.14624594151973724,
      "eval_runtime": 404.3759,
      "eval_samples_per_second": 97.115,
      "eval_steps_per_second": 24.279,
      "num_input_tokens_seen": 37954104,
      "step": 23600
    },
    {
      "epoch": 1.0685846216528214,
      "grad_norm": 0.39234593510627747,
      "learning_rate": 1.801768056436429e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 37962744,
      "step": 23605
    },
    {
      "epoch": 1.0688109735394644,
      "grad_norm": 0.4828003942966461,
      "learning_rate": 1.8008254302718035e-05,
      "loss": 0.0948,
      "num_input_tokens_seen": 37970776,
      "step": 23610
    },
    {
      "epoch": 1.0690373254261074,
      "grad_norm": 0.7222295999526978,
      "learning_rate": 1.7998829119286837e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 37978712,
      "step": 23615
    },
    {
      "epoch": 1.0692636773127504,
      "grad_norm": 0.8038733601570129,
      "learning_rate": 1.798940501552418e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 37987608,
      "step": 23620
    },
    {
      "epoch": 1.0694900291993934,
      "grad_norm": 0.49765703082084656,
      "learning_rate": 1.797998199288336e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 37995448,
      "step": 23625
    },
    {
      "epoch": 1.0697163810860364,
      "grad_norm": 0.5875737071037292,
      "learning_rate": 1.7970560052817543e-05,
      "loss": 0.1074,
      "num_input_tokens_seen": 38003256,
      "step": 23630
    },
    {
      "epoch": 1.0699427329726794,
      "grad_norm": 0.7496333122253418,
      "learning_rate": 1.7961139196779702e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 38011064,
      "step": 23635
    },
    {
      "epoch": 1.0701690848593224,
      "grad_norm": 0.9535430073738098,
      "learning_rate": 1.7951719426222647e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 38019096,
      "step": 23640
    },
    {
      "epoch": 1.0703954367459654,
      "grad_norm": 0.5619316101074219,
      "learning_rate": 1.794230074259904e-05,
      "loss": 0.1895,
      "num_input_tokens_seen": 38027384,
      "step": 23645
    },
    {
      "epoch": 1.0706217886326082,
      "grad_norm": 1.8445618152618408,
      "learning_rate": 1.7932883147361336e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 38035576,
      "step": 23650
    },
    {
      "epoch": 1.0708481405192511,
      "grad_norm": 1.0835310220718384,
      "learning_rate": 1.7923466641961865e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 38043512,
      "step": 23655
    },
    {
      "epoch": 1.0710744924058941,
      "grad_norm": 0.5110387802124023,
      "learning_rate": 1.791405122785278e-05,
      "loss": 0.1706,
      "num_input_tokens_seen": 38051608,
      "step": 23660
    },
    {
      "epoch": 1.0713008442925371,
      "grad_norm": 0.34428250789642334,
      "learning_rate": 1.7904636906486037e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 38059288,
      "step": 23665
    },
    {
      "epoch": 1.0715271961791801,
      "grad_norm": 0.6928130984306335,
      "learning_rate": 1.7895223679313448e-05,
      "loss": 0.1725,
      "num_input_tokens_seen": 38067256,
      "step": 23670
    },
    {
      "epoch": 1.0717535480658231,
      "grad_norm": 0.5128175616264343,
      "learning_rate": 1.7885811547786653e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 38074648,
      "step": 23675
    },
    {
      "epoch": 1.0719798999524661,
      "grad_norm": 0.4901770055294037,
      "learning_rate": 1.7876400513357115e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 38082232,
      "step": 23680
    },
    {
      "epoch": 1.0722062518391091,
      "grad_norm": 0.499774307012558,
      "learning_rate": 1.7866990577476146e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 38089912,
      "step": 23685
    },
    {
      "epoch": 1.072432603725752,
      "grad_norm": 0.9254423975944519,
      "learning_rate": 1.7857581741594863e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 38098008,
      "step": 23690
    },
    {
      "epoch": 1.072658955612395,
      "grad_norm": 0.6072790026664734,
      "learning_rate": 1.7848174007164237e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 38105752,
      "step": 23695
    },
    {
      "epoch": 1.072885307499038,
      "grad_norm": 0.5322611331939697,
      "learning_rate": 1.7838767375635052e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 38113528,
      "step": 23700
    },
    {
      "epoch": 1.073111659385681,
      "grad_norm": 1.4042060375213623,
      "learning_rate": 1.782936184845793e-05,
      "loss": 0.1932,
      "num_input_tokens_seen": 38121464,
      "step": 23705
    },
    {
      "epoch": 1.0733380112723239,
      "grad_norm": 0.4308897852897644,
      "learning_rate": 1.7819957427083334e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 38128984,
      "step": 23710
    },
    {
      "epoch": 1.0735643631589669,
      "grad_norm": 0.5043941736221313,
      "learning_rate": 1.7810554112961516e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 38137048,
      "step": 23715
    },
    {
      "epoch": 1.0737907150456099,
      "grad_norm": 0.6128694415092468,
      "learning_rate": 1.7801151907542607e-05,
      "loss": 0.1042,
      "num_input_tokens_seen": 38145144,
      "step": 23720
    },
    {
      "epoch": 1.0740170669322529,
      "grad_norm": 0.720808207988739,
      "learning_rate": 1.7791750812276547e-05,
      "loss": 0.1401,
      "num_input_tokens_seen": 38153016,
      "step": 23725
    },
    {
      "epoch": 1.0742434188188958,
      "grad_norm": 0.7569738030433655,
      "learning_rate": 1.778235082861309e-05,
      "loss": 0.1188,
      "num_input_tokens_seen": 38160504,
      "step": 23730
    },
    {
      "epoch": 1.0744697707055388,
      "grad_norm": 1.342305302619934,
      "learning_rate": 1.777295195800184e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 38168280,
      "step": 23735
    },
    {
      "epoch": 1.0746961225921818,
      "grad_norm": 0.3408253788948059,
      "learning_rate": 1.7763554201892215e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 38176280,
      "step": 23740
    },
    {
      "epoch": 1.0749224744788248,
      "grad_norm": 0.4196633994579315,
      "learning_rate": 1.7754157561733476e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 38184408,
      "step": 23745
    },
    {
      "epoch": 1.0751488263654678,
      "grad_norm": 0.38194504380226135,
      "learning_rate": 1.7744762038974702e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 38192568,
      "step": 23750
    },
    {
      "epoch": 1.0753751782521108,
      "grad_norm": 0.39205867052078247,
      "learning_rate": 1.7735367635064788e-05,
      "loss": 0.1045,
      "num_input_tokens_seen": 38199928,
      "step": 23755
    },
    {
      "epoch": 1.0756015301387536,
      "grad_norm": 0.812910258769989,
      "learning_rate": 1.7725974351452474e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 38207576,
      "step": 23760
    },
    {
      "epoch": 1.0758278820253966,
      "grad_norm": 0.6936744451522827,
      "learning_rate": 1.771658218958634e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 38215128,
      "step": 23765
    },
    {
      "epoch": 1.0760542339120396,
      "grad_norm": 0.8027763366699219,
      "learning_rate": 1.770719115091475e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 38223096,
      "step": 23770
    },
    {
      "epoch": 1.0762805857986826,
      "grad_norm": 0.37699657678604126,
      "learning_rate": 1.7697801236885935e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 38232280,
      "step": 23775
    },
    {
      "epoch": 1.0765069376853256,
      "grad_norm": 0.5417057275772095,
      "learning_rate": 1.7688412448947944e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 38240376,
      "step": 23780
    },
    {
      "epoch": 1.0767332895719686,
      "grad_norm": 0.6931489706039429,
      "learning_rate": 1.767902478854862e-05,
      "loss": 0.1736,
      "num_input_tokens_seen": 38248152,
      "step": 23785
    },
    {
      "epoch": 1.0769596414586116,
      "grad_norm": 0.43804001808166504,
      "learning_rate": 1.766963825713569e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 38256056,
      "step": 23790
    },
    {
      "epoch": 1.0771859933452546,
      "grad_norm": 1.1299172639846802,
      "learning_rate": 1.766025285615665e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 38263864,
      "step": 23795
    },
    {
      "epoch": 1.0774123452318976,
      "grad_norm": 0.9197471737861633,
      "learning_rate": 1.7650868587058854e-05,
      "loss": 0.1638,
      "num_input_tokens_seen": 38272056,
      "step": 23800
    },
    {
      "epoch": 1.0774123452318976,
      "eval_loss": 0.14594325423240662,
      "eval_runtime": 404.6101,
      "eval_samples_per_second": 97.059,
      "eval_steps_per_second": 24.265,
      "num_input_tokens_seen": 38272056,
      "step": 23800
    },
    {
      "epoch": 1.0776386971185405,
      "grad_norm": 1.3460633754730225,
      "learning_rate": 1.7641485451289484e-05,
      "loss": 0.1642,
      "num_input_tokens_seen": 38280312,
      "step": 23805
    },
    {
      "epoch": 1.0778650490051835,
      "grad_norm": 0.49555760622024536,
      "learning_rate": 1.7632103450295534e-05,
      "loss": 0.1884,
      "num_input_tokens_seen": 38288280,
      "step": 23810
    },
    {
      "epoch": 1.0780914008918265,
      "grad_norm": 0.35183459520339966,
      "learning_rate": 1.762272258552381e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 38296664,
      "step": 23815
    },
    {
      "epoch": 1.0783177527784693,
      "grad_norm": 0.8009275197982788,
      "learning_rate": 1.7613342858420988e-05,
      "loss": 0.1636,
      "num_input_tokens_seen": 38304856,
      "step": 23820
    },
    {
      "epoch": 1.0785441046651123,
      "grad_norm": 1.2713133096694946,
      "learning_rate": 1.760396427043351e-05,
      "loss": 0.1747,
      "num_input_tokens_seen": 38313112,
      "step": 23825
    },
    {
      "epoch": 1.0787704565517553,
      "grad_norm": 0.5590147376060486,
      "learning_rate": 1.7594586823007696e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 38321176,
      "step": 23830
    },
    {
      "epoch": 1.0789968084383983,
      "grad_norm": 0.7459941506385803,
      "learning_rate": 1.7585210517589646e-05,
      "loss": 0.1147,
      "num_input_tokens_seen": 38330040,
      "step": 23835
    },
    {
      "epoch": 1.0792231603250413,
      "grad_norm": 0.8209688067436218,
      "learning_rate": 1.7575835355625314e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 38337944,
      "step": 23840
    },
    {
      "epoch": 1.0794495122116843,
      "grad_norm": 1.327454686164856,
      "learning_rate": 1.756646133856048e-05,
      "loss": 0.1733,
      "num_input_tokens_seen": 38345912,
      "step": 23845
    },
    {
      "epoch": 1.0796758640983273,
      "grad_norm": 1.1145988702774048,
      "learning_rate": 1.7557088467840714e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 38353784,
      "step": 23850
    },
    {
      "epoch": 1.0799022159849703,
      "grad_norm": 0.28226280212402344,
      "learning_rate": 1.7547716744911438e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 38361528,
      "step": 23855
    },
    {
      "epoch": 1.0801285678716133,
      "grad_norm": 0.698265016078949,
      "learning_rate": 1.7538346171217902e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 38369784,
      "step": 23860
    },
    {
      "epoch": 1.0803549197582563,
      "grad_norm": 0.6298686861991882,
      "learning_rate": 1.7528976748205146e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 38376984,
      "step": 23865
    },
    {
      "epoch": 1.0805812716448993,
      "grad_norm": 0.553532600402832,
      "learning_rate": 1.751960847731807e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 38385272,
      "step": 23870
    },
    {
      "epoch": 1.080807623531542,
      "grad_norm": 0.5806223154067993,
      "learning_rate": 1.7510241360001362e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 38392888,
      "step": 23875
    },
    {
      "epoch": 1.081033975418185,
      "grad_norm": 0.9217005372047424,
      "learning_rate": 1.7500875397699562e-05,
      "loss": 0.1916,
      "num_input_tokens_seen": 38400504,
      "step": 23880
    },
    {
      "epoch": 1.081260327304828,
      "grad_norm": 0.4959612190723419,
      "learning_rate": 1.7491510591857015e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 38408920,
      "step": 23885
    },
    {
      "epoch": 1.081486679191471,
      "grad_norm": 0.4759417772293091,
      "learning_rate": 1.7482146943917896e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 38416600,
      "step": 23890
    },
    {
      "epoch": 1.081713031078114,
      "grad_norm": 0.467758446931839,
      "learning_rate": 1.7472784455326185e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 38424600,
      "step": 23895
    },
    {
      "epoch": 1.081939382964757,
      "grad_norm": 0.7870258688926697,
      "learning_rate": 1.746342312752572e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 38432696,
      "step": 23900
    },
    {
      "epoch": 1.0821657348514,
      "grad_norm": 0.2389173060655594,
      "learning_rate": 1.74540629619601e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 38440632,
      "step": 23905
    },
    {
      "epoch": 1.082392086738043,
      "grad_norm": 0.4318406581878662,
      "learning_rate": 1.7444703960072815e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 38448472,
      "step": 23910
    },
    {
      "epoch": 1.082618438624686,
      "grad_norm": 0.8938871622085571,
      "learning_rate": 1.7435346123307118e-05,
      "loss": 0.167,
      "num_input_tokens_seen": 38455960,
      "step": 23915
    },
    {
      "epoch": 1.082844790511329,
      "grad_norm": 1.2327989339828491,
      "learning_rate": 1.742598945310611e-05,
      "loss": 0.1405,
      "num_input_tokens_seen": 38463800,
      "step": 23920
    },
    {
      "epoch": 1.083071142397972,
      "grad_norm": 0.604022204875946,
      "learning_rate": 1.741663395091272e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 38472312,
      "step": 23925
    },
    {
      "epoch": 1.083297494284615,
      "grad_norm": 1.138383150100708,
      "learning_rate": 1.7407279618169657e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 38480312,
      "step": 23930
    },
    {
      "epoch": 1.0835238461712577,
      "grad_norm": 0.5985109806060791,
      "learning_rate": 1.73979264563195e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 38489240,
      "step": 23935
    },
    {
      "epoch": 1.0837501980579007,
      "grad_norm": 0.6540613174438477,
      "learning_rate": 1.7388574466804625e-05,
      "loss": 0.128,
      "num_input_tokens_seen": 38497688,
      "step": 23940
    },
    {
      "epoch": 1.0839765499445437,
      "grad_norm": 0.4989885091781616,
      "learning_rate": 1.7379223651067207e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 38505464,
      "step": 23945
    },
    {
      "epoch": 1.0842029018311867,
      "grad_norm": 0.6960896849632263,
      "learning_rate": 1.736987401054928e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 38513624,
      "step": 23950
    },
    {
      "epoch": 1.0844292537178297,
      "grad_norm": 0.5938047766685486,
      "learning_rate": 1.736052554669266e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 38521560,
      "step": 23955
    },
    {
      "epoch": 1.0846556056044727,
      "grad_norm": 0.3510644733905792,
      "learning_rate": 1.7351178260939007e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 38529528,
      "step": 23960
    },
    {
      "epoch": 1.0848819574911157,
      "grad_norm": 0.6388384103775024,
      "learning_rate": 1.7341832154729794e-05,
      "loss": 0.1512,
      "num_input_tokens_seen": 38537336,
      "step": 23965
    },
    {
      "epoch": 1.0851083093777587,
      "grad_norm": 1.1248809099197388,
      "learning_rate": 1.7332487229506286e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 38545112,
      "step": 23970
    },
    {
      "epoch": 1.0853346612644017,
      "grad_norm": 0.8720222115516663,
      "learning_rate": 1.732314348670961e-05,
      "loss": 0.1064,
      "num_input_tokens_seen": 38554456,
      "step": 23975
    },
    {
      "epoch": 1.0855610131510447,
      "grad_norm": 0.6524847745895386,
      "learning_rate": 1.7313800927780686e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 38562136,
      "step": 23980
    },
    {
      "epoch": 1.0857873650376877,
      "grad_norm": 0.5566949248313904,
      "learning_rate": 1.7304459554160245e-05,
      "loss": 0.1129,
      "num_input_tokens_seen": 38570744,
      "step": 23985
    },
    {
      "epoch": 1.0860137169243305,
      "grad_norm": 0.8628884553909302,
      "learning_rate": 1.7295119367288853e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 38578552,
      "step": 23990
    },
    {
      "epoch": 1.0862400688109735,
      "grad_norm": 0.4707282483577728,
      "learning_rate": 1.728578036860688e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 38587128,
      "step": 23995
    },
    {
      "epoch": 1.0864664206976165,
      "grad_norm": 0.715234100818634,
      "learning_rate": 1.7276442559554513e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 38594968,
      "step": 24000
    },
    {
      "epoch": 1.0864664206976165,
      "eval_loss": 0.14537878334522247,
      "eval_runtime": 405.1509,
      "eval_samples_per_second": 96.929,
      "eval_steps_per_second": 24.233,
      "num_input_tokens_seen": 38594968,
      "step": 24000
    },
    {
      "epoch": 1.0866927725842594,
      "grad_norm": 0.6137085556983948,
      "learning_rate": 1.726710594157177e-05,
      "loss": 0.1055,
      "num_input_tokens_seen": 38602936,
      "step": 24005
    },
    {
      "epoch": 1.0869191244709024,
      "grad_norm": 0.5691301822662354,
      "learning_rate": 1.725777051609846e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 38610392,
      "step": 24010
    },
    {
      "epoch": 1.0871454763575454,
      "grad_norm": 0.5876774787902832,
      "learning_rate": 1.7248436284574228e-05,
      "loss": 0.1688,
      "num_input_tokens_seen": 38618264,
      "step": 24015
    },
    {
      "epoch": 1.0873718282441884,
      "grad_norm": 0.7158837914466858,
      "learning_rate": 1.723910324843855e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 38626040,
      "step": 24020
    },
    {
      "epoch": 1.0875981801308314,
      "grad_norm": 0.6911791563034058,
      "learning_rate": 1.722977140913067e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 38634328,
      "step": 24025
    },
    {
      "epoch": 1.0878245320174744,
      "grad_norm": 0.4800361692905426,
      "learning_rate": 1.7220440768089688e-05,
      "loss": 0.1811,
      "num_input_tokens_seen": 38642296,
      "step": 24030
    },
    {
      "epoch": 1.0880508839041174,
      "grad_norm": 0.5580857396125793,
      "learning_rate": 1.7211111326754505e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 38650904,
      "step": 24035
    },
    {
      "epoch": 1.0882772357907604,
      "grad_norm": 0.5393019914627075,
      "learning_rate": 1.720178308656383e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 38658488,
      "step": 24040
    },
    {
      "epoch": 1.0885035876774034,
      "grad_norm": 0.7183153033256531,
      "learning_rate": 1.719245604895621e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 38666328,
      "step": 24045
    },
    {
      "epoch": 1.0887299395640462,
      "grad_norm": 1.1594829559326172,
      "learning_rate": 1.7183130215369972e-05,
      "loss": 0.1639,
      "num_input_tokens_seen": 38674392,
      "step": 24050
    },
    {
      "epoch": 1.0889562914506892,
      "grad_norm": 0.5764090418815613,
      "learning_rate": 1.7173805587243292e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 38681784,
      "step": 24055
    },
    {
      "epoch": 1.0891826433373322,
      "grad_norm": 0.34514638781547546,
      "learning_rate": 1.7164482166014147e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 38690424,
      "step": 24060
    },
    {
      "epoch": 1.0894089952239752,
      "grad_norm": 0.8195791244506836,
      "learning_rate": 1.7155159953120313e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 38698456,
      "step": 24065
    },
    {
      "epoch": 1.0896353471106182,
      "grad_norm": 0.39232879877090454,
      "learning_rate": 1.714583894999941e-05,
      "loss": 0.0918,
      "num_input_tokens_seen": 38706520,
      "step": 24070
    },
    {
      "epoch": 1.0898616989972612,
      "grad_norm": 0.5185642838478088,
      "learning_rate": 1.7136519158088826e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 38714520,
      "step": 24075
    },
    {
      "epoch": 1.0900880508839041,
      "grad_norm": 1.9268953800201416,
      "learning_rate": 1.712720057882581e-05,
      "loss": 0.1558,
      "num_input_tokens_seen": 38723384,
      "step": 24080
    },
    {
      "epoch": 1.0903144027705471,
      "grad_norm": 0.6486467123031616,
      "learning_rate": 1.7117883213647413e-05,
      "loss": 0.1509,
      "num_input_tokens_seen": 38732056,
      "step": 24085
    },
    {
      "epoch": 1.0905407546571901,
      "grad_norm": 0.4155331552028656,
      "learning_rate": 1.710856706399046e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 38739832,
      "step": 24090
    },
    {
      "epoch": 1.0907671065438331,
      "grad_norm": 0.8519158363342285,
      "learning_rate": 1.7099252131291648e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 38747704,
      "step": 24095
    },
    {
      "epoch": 1.090993458430476,
      "grad_norm": 0.6991653442382812,
      "learning_rate": 1.708993841698744e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 38755480,
      "step": 24100
    },
    {
      "epoch": 1.091219810317119,
      "grad_norm": 0.415019690990448,
      "learning_rate": 1.7080625922514132e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 38763160,
      "step": 24105
    },
    {
      "epoch": 1.091446162203762,
      "grad_norm": 1.010732650756836,
      "learning_rate": 1.7071314649307836e-05,
      "loss": 0.1608,
      "num_input_tokens_seen": 38770904,
      "step": 24110
    },
    {
      "epoch": 1.0916725140904049,
      "grad_norm": 0.7775738835334778,
      "learning_rate": 1.7062004598804448e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 38779032,
      "step": 24115
    },
    {
      "epoch": 1.0918988659770479,
      "grad_norm": 0.5163336396217346,
      "learning_rate": 1.7052695772439702e-05,
      "loss": 0.1215,
      "num_input_tokens_seen": 38786968,
      "step": 24120
    },
    {
      "epoch": 1.0921252178636909,
      "grad_norm": 0.2853586971759796,
      "learning_rate": 1.7043388171649154e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 38795480,
      "step": 24125
    },
    {
      "epoch": 1.0923515697503339,
      "grad_norm": 0.5065311193466187,
      "learning_rate": 1.7034081797868127e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 38804056,
      "step": 24130
    },
    {
      "epoch": 1.0925779216369769,
      "grad_norm": 1.461044430732727,
      "learning_rate": 1.70247766525318e-05,
      "loss": 0.1125,
      "num_input_tokens_seen": 38812344,
      "step": 24135
    },
    {
      "epoch": 1.0928042735236199,
      "grad_norm": 0.42390623688697815,
      "learning_rate": 1.701547273707514e-05,
      "loss": 0.142,
      "num_input_tokens_seen": 38820504,
      "step": 24140
    },
    {
      "epoch": 1.0930306254102629,
      "grad_norm": 1.0499205589294434,
      "learning_rate": 1.7006170052932916e-05,
      "loss": 0.1463,
      "num_input_tokens_seen": 38828984,
      "step": 24145
    },
    {
      "epoch": 1.0932569772969059,
      "grad_norm": 0.41199976205825806,
      "learning_rate": 1.6996868601539735e-05,
      "loss": 0.1109,
      "num_input_tokens_seen": 38837112,
      "step": 24150
    },
    {
      "epoch": 1.0934833291835488,
      "grad_norm": 0.4057564437389374,
      "learning_rate": 1.6987568384329977e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 38844920,
      "step": 24155
    },
    {
      "epoch": 1.0937096810701916,
      "grad_norm": 0.41325023770332336,
      "learning_rate": 1.6978269402737866e-05,
      "loss": 0.1441,
      "num_input_tokens_seen": 38852600,
      "step": 24160
    },
    {
      "epoch": 1.0939360329568346,
      "grad_norm": 0.3550783395767212,
      "learning_rate": 1.696897165819743e-05,
      "loss": 0.1149,
      "num_input_tokens_seen": 38860984,
      "step": 24165
    },
    {
      "epoch": 1.0941623848434776,
      "grad_norm": 1.2823163270950317,
      "learning_rate": 1.6959675152142487e-05,
      "loss": 0.196,
      "num_input_tokens_seen": 38868792,
      "step": 24170
    },
    {
      "epoch": 1.0943887367301206,
      "grad_norm": 0.9182760715484619,
      "learning_rate": 1.6950379886006667e-05,
      "loss": 0.1633,
      "num_input_tokens_seen": 38877080,
      "step": 24175
    },
    {
      "epoch": 1.0946150886167636,
      "grad_norm": 0.8374834060668945,
      "learning_rate": 1.6941085861223438e-05,
      "loss": 0.1662,
      "num_input_tokens_seen": 38885592,
      "step": 24180
    },
    {
      "epoch": 1.0948414405034066,
      "grad_norm": 1.6161012649536133,
      "learning_rate": 1.6931793079226034e-05,
      "loss": 0.1334,
      "num_input_tokens_seen": 38893304,
      "step": 24185
    },
    {
      "epoch": 1.0950677923900496,
      "grad_norm": 0.4811707139015198,
      "learning_rate": 1.692250154144754e-05,
      "loss": 0.1189,
      "num_input_tokens_seen": 38901944,
      "step": 24190
    },
    {
      "epoch": 1.0952941442766926,
      "grad_norm": 0.5784329771995544,
      "learning_rate": 1.6913211249320807e-05,
      "loss": 0.1181,
      "num_input_tokens_seen": 38909816,
      "step": 24195
    },
    {
      "epoch": 1.0955204961633356,
      "grad_norm": 0.6911270022392273,
      "learning_rate": 1.6903922204278522e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 38918264,
      "step": 24200
    },
    {
      "epoch": 1.0955204961633356,
      "eval_loss": 0.14565107226371765,
      "eval_runtime": 405.0877,
      "eval_samples_per_second": 96.944,
      "eval_steps_per_second": 24.237,
      "num_input_tokens_seen": 38918264,
      "step": 24200
    },
    {
      "epoch": 1.0957468480499786,
      "grad_norm": 0.6189802289009094,
      "learning_rate": 1.6894634407753186e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 38926520,
      "step": 24205
    },
    {
      "epoch": 1.0959731999366216,
      "grad_norm": 0.41496825218200684,
      "learning_rate": 1.6885347861177077e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 38935224,
      "step": 24210
    },
    {
      "epoch": 1.0961995518232643,
      "grad_norm": 1.3387870788574219,
      "learning_rate": 1.6876062565982298e-05,
      "loss": 0.1569,
      "num_input_tokens_seen": 38943480,
      "step": 24215
    },
    {
      "epoch": 1.0964259037099073,
      "grad_norm": 0.4625508189201355,
      "learning_rate": 1.6866778523600774e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 38951768,
      "step": 24220
    },
    {
      "epoch": 1.0966522555965503,
      "grad_norm": 0.7262976169586182,
      "learning_rate": 1.6857495735464195e-05,
      "loss": 0.1476,
      "num_input_tokens_seen": 38960152,
      "step": 24225
    },
    {
      "epoch": 1.0968786074831933,
      "grad_norm": 0.7585103511810303,
      "learning_rate": 1.6848214203004115e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 38968312,
      "step": 24230
    },
    {
      "epoch": 1.0971049593698363,
      "grad_norm": 0.5368009805679321,
      "learning_rate": 1.6838933927651835e-05,
      "loss": 0.095,
      "num_input_tokens_seen": 38976952,
      "step": 24235
    },
    {
      "epoch": 1.0973313112564793,
      "grad_norm": 0.991020143032074,
      "learning_rate": 1.6829654910838506e-05,
      "loss": 0.1468,
      "num_input_tokens_seen": 38984504,
      "step": 24240
    },
    {
      "epoch": 1.0975576631431223,
      "grad_norm": 0.3805496394634247,
      "learning_rate": 1.6820377153995065e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 38993080,
      "step": 24245
    },
    {
      "epoch": 1.0977840150297653,
      "grad_norm": 0.8438015580177307,
      "learning_rate": 1.681110065855226e-05,
      "loss": 0.1288,
      "num_input_tokens_seen": 39000664,
      "step": 24250
    },
    {
      "epoch": 1.0980103669164083,
      "grad_norm": 0.5249488949775696,
      "learning_rate": 1.6801825425940642e-05,
      "loss": 0.1178,
      "num_input_tokens_seen": 39008632,
      "step": 24255
    },
    {
      "epoch": 1.0982367188030513,
      "grad_norm": 1.1305955648422241,
      "learning_rate": 1.679255145759056e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 39016952,
      "step": 24260
    },
    {
      "epoch": 1.0984630706896943,
      "grad_norm": 0.3678140342235565,
      "learning_rate": 1.6783278754932187e-05,
      "loss": 0.1445,
      "num_input_tokens_seen": 39024664,
      "step": 24265
    },
    {
      "epoch": 1.0986894225763373,
      "grad_norm": 0.8418135046958923,
      "learning_rate": 1.6774007319395496e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 39033144,
      "step": 24270
    },
    {
      "epoch": 1.09891577446298,
      "grad_norm": 1.5160826444625854,
      "learning_rate": 1.6764737152410243e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 39040600,
      "step": 24275
    },
    {
      "epoch": 1.099142126349623,
      "grad_norm": 0.8084892630577087,
      "learning_rate": 1.6755468255406016e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 39048248,
      "step": 24280
    },
    {
      "epoch": 1.099368478236266,
      "grad_norm": 0.5722837448120117,
      "learning_rate": 1.674620062981219e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 39055928,
      "step": 24285
    },
    {
      "epoch": 1.099594830122909,
      "grad_norm": 0.35533663630485535,
      "learning_rate": 1.6736934277057947e-05,
      "loss": 0.1213,
      "num_input_tokens_seen": 39064056,
      "step": 24290
    },
    {
      "epoch": 1.099821182009552,
      "grad_norm": 0.5125505924224854,
      "learning_rate": 1.6727669198572286e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 39072760,
      "step": 24295
    },
    {
      "epoch": 1.100047533896195,
      "grad_norm": 0.4177374839782715,
      "learning_rate": 1.6718405395783984e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 39080376,
      "step": 24300
    },
    {
      "epoch": 1.100273885782838,
      "grad_norm": 0.844732940196991,
      "learning_rate": 1.6709142870121643e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 39088600,
      "step": 24305
    },
    {
      "epoch": 1.100500237669481,
      "grad_norm": 1.1841920614242554,
      "learning_rate": 1.669988162301367e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 39096728,
      "step": 24310
    },
    {
      "epoch": 1.100726589556124,
      "grad_norm": 1.0255378484725952,
      "learning_rate": 1.6690621655888243e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 39104792,
      "step": 24315
    },
    {
      "epoch": 1.100952941442767,
      "grad_norm": 0.9256148934364319,
      "learning_rate": 1.6681362970173386e-05,
      "loss": 0.1594,
      "num_input_tokens_seen": 39112760,
      "step": 24320
    },
    {
      "epoch": 1.10117929332941,
      "grad_norm": 0.5914594531059265,
      "learning_rate": 1.6672105567296904e-05,
      "loss": 0.1712,
      "num_input_tokens_seen": 39120888,
      "step": 24325
    },
    {
      "epoch": 1.1014056452160528,
      "grad_norm": 0.6077331900596619,
      "learning_rate": 1.666284944868639e-05,
      "loss": 0.1191,
      "num_input_tokens_seen": 39128920,
      "step": 24330
    },
    {
      "epoch": 1.1016319971026958,
      "grad_norm": 1.057590126991272,
      "learning_rate": 1.665359461576927e-05,
      "loss": 0.2159,
      "num_input_tokens_seen": 39137432,
      "step": 24335
    },
    {
      "epoch": 1.1018583489893388,
      "grad_norm": 0.6637601256370544,
      "learning_rate": 1.6644341069972736e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 39146872,
      "step": 24340
    },
    {
      "epoch": 1.1020847008759818,
      "grad_norm": 0.667482852935791,
      "learning_rate": 1.6635088812723813e-05,
      "loss": 0.1408,
      "num_input_tokens_seen": 39155512,
      "step": 24345
    },
    {
      "epoch": 1.1023110527626248,
      "grad_norm": 0.7134560346603394,
      "learning_rate": 1.6625837845449328e-05,
      "loss": 0.1834,
      "num_input_tokens_seen": 39163512,
      "step": 24350
    },
    {
      "epoch": 1.1025374046492677,
      "grad_norm": 0.6467038989067078,
      "learning_rate": 1.6616588169575874e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 39171480,
      "step": 24355
    },
    {
      "epoch": 1.1027637565359107,
      "grad_norm": 0.6217794418334961,
      "learning_rate": 1.6607339786529878e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 39179736,
      "step": 24360
    },
    {
      "epoch": 1.1029901084225537,
      "grad_norm": 0.3990156054496765,
      "learning_rate": 1.659809269773756e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 39187896,
      "step": 24365
    },
    {
      "epoch": 1.1032164603091967,
      "grad_norm": 0.7896202206611633,
      "learning_rate": 1.658884690462493e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 39195736,
      "step": 24370
    },
    {
      "epoch": 1.1034428121958397,
      "grad_norm": 0.5892680287361145,
      "learning_rate": 1.6579602408617813e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 39203064,
      "step": 24375
    },
    {
      "epoch": 1.1036691640824827,
      "grad_norm": 0.8601559400558472,
      "learning_rate": 1.657035921114181e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 39210488,
      "step": 24380
    },
    {
      "epoch": 1.1038955159691257,
      "grad_norm": 0.7074025273323059,
      "learning_rate": 1.656111731362236e-05,
      "loss": 0.1172,
      "num_input_tokens_seen": 39218328,
      "step": 24385
    },
    {
      "epoch": 1.1041218678557685,
      "grad_norm": 1.7261552810668945,
      "learning_rate": 1.6551876717484666e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 39226744,
      "step": 24390
    },
    {
      "epoch": 1.1043482197424115,
      "grad_norm": 0.4338149130344391,
      "learning_rate": 1.6542637424153752e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 39234520,
      "step": 24395
    },
    {
      "epoch": 1.1045745716290545,
      "grad_norm": 0.4587666988372803,
      "learning_rate": 1.6533399435054418e-05,
      "loss": 0.1375,
      "num_input_tokens_seen": 39242584,
      "step": 24400
    },
    {
      "epoch": 1.1045745716290545,
      "eval_loss": 0.14528888463974,
      "eval_runtime": 405.1507,
      "eval_samples_per_second": 96.929,
      "eval_steps_per_second": 24.233,
      "num_input_tokens_seen": 39242584,
      "step": 24400
    },
    {
      "epoch": 1.1048009235156975,
      "grad_norm": 0.6246293187141418,
      "learning_rate": 1.6524162751611304e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 39250360,
      "step": 24405
    },
    {
      "epoch": 1.1050272754023405,
      "grad_norm": 0.5411859154701233,
      "learning_rate": 1.6514927375248796e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 39257592,
      "step": 24410
    },
    {
      "epoch": 1.1052536272889835,
      "grad_norm": 0.538102924823761,
      "learning_rate": 1.6505693307391127e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 39265848,
      "step": 24415
    },
    {
      "epoch": 1.1054799791756265,
      "grad_norm": 0.4030859172344208,
      "learning_rate": 1.6496460549462288e-05,
      "loss": 0.0978,
      "num_input_tokens_seen": 39273400,
      "step": 24420
    },
    {
      "epoch": 1.1057063310622695,
      "grad_norm": 0.4270245432853699,
      "learning_rate": 1.6487229102886097e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 39281240,
      "step": 24425
    },
    {
      "epoch": 1.1059326829489124,
      "grad_norm": 0.7165495157241821,
      "learning_rate": 1.6477998969086155e-05,
      "loss": 0.1185,
      "num_input_tokens_seen": 39289368,
      "step": 24430
    },
    {
      "epoch": 1.1061590348355554,
      "grad_norm": 0.37078505754470825,
      "learning_rate": 1.646877014948587e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 39297112,
      "step": 24435
    },
    {
      "epoch": 1.1063853867221982,
      "grad_norm": 0.8893968462944031,
      "learning_rate": 1.6459542645508433e-05,
      "loss": 0.0842,
      "num_input_tokens_seen": 39305464,
      "step": 24440
    },
    {
      "epoch": 1.1066117386088412,
      "grad_norm": 0.7594598531723022,
      "learning_rate": 1.6450316458576852e-05,
      "loss": 0.1352,
      "num_input_tokens_seen": 39313336,
      "step": 24445
    },
    {
      "epoch": 1.1068380904954842,
      "grad_norm": 0.8689106106758118,
      "learning_rate": 1.6441091590113912e-05,
      "loss": 0.1634,
      "num_input_tokens_seen": 39321048,
      "step": 24450
    },
    {
      "epoch": 1.1070644423821272,
      "grad_norm": 0.37106406688690186,
      "learning_rate": 1.6431868041542213e-05,
      "loss": 0.1183,
      "num_input_tokens_seen": 39329272,
      "step": 24455
    },
    {
      "epoch": 1.1072907942687702,
      "grad_norm": 0.602191150188446,
      "learning_rate": 1.6422645814284123e-05,
      "loss": 0.1609,
      "num_input_tokens_seen": 39337112,
      "step": 24460
    },
    {
      "epoch": 1.1075171461554132,
      "grad_norm": 0.5258868932723999,
      "learning_rate": 1.6413424909761846e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 39345176,
      "step": 24465
    },
    {
      "epoch": 1.1077434980420562,
      "grad_norm": 0.7086993455886841,
      "learning_rate": 1.640420532939736e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 39353112,
      "step": 24470
    },
    {
      "epoch": 1.1079698499286992,
      "grad_norm": 0.4737364947795868,
      "learning_rate": 1.639498707461242e-05,
      "loss": 0.1082,
      "num_input_tokens_seen": 39361176,
      "step": 24475
    },
    {
      "epoch": 1.1081962018153422,
      "grad_norm": 0.9770500659942627,
      "learning_rate": 1.6385770146828614e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 39369144,
      "step": 24480
    },
    {
      "epoch": 1.1084225537019852,
      "grad_norm": 0.6629988551139832,
      "learning_rate": 1.637655454746731e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 39376984,
      "step": 24485
    },
    {
      "epoch": 1.1086489055886282,
      "grad_norm": 0.39158347249031067,
      "learning_rate": 1.6367340277949658e-05,
      "loss": 0.1407,
      "num_input_tokens_seen": 39385176,
      "step": 24490
    },
    {
      "epoch": 1.1088752574752712,
      "grad_norm": 0.8737599849700928,
      "learning_rate": 1.635812733969663e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 39393432,
      "step": 24495
    },
    {
      "epoch": 1.109101609361914,
      "grad_norm": 1.0771790742874146,
      "learning_rate": 1.634891573412896e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 39401368,
      "step": 24500
    },
    {
      "epoch": 1.109327961248557,
      "grad_norm": 0.8042096495628357,
      "learning_rate": 1.6339705462667196e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 39409240,
      "step": 24505
    },
    {
      "epoch": 1.1095543131352,
      "grad_norm": 0.49897921085357666,
      "learning_rate": 1.633049652673169e-05,
      "loss": 0.1124,
      "num_input_tokens_seen": 39417624,
      "step": 24510
    },
    {
      "epoch": 1.109780665021843,
      "grad_norm": 0.9570324420928955,
      "learning_rate": 1.632128892774256e-05,
      "loss": 0.108,
      "num_input_tokens_seen": 39425816,
      "step": 24515
    },
    {
      "epoch": 1.110007016908486,
      "grad_norm": 0.9478817582130432,
      "learning_rate": 1.6312082667119737e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 39434168,
      "step": 24520
    },
    {
      "epoch": 1.110233368795129,
      "grad_norm": 0.7100887894630432,
      "learning_rate": 1.630287774628296e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 39443256,
      "step": 24525
    },
    {
      "epoch": 1.110459720681772,
      "grad_norm": 0.7320757508277893,
      "learning_rate": 1.6293674166651718e-05,
      "loss": 0.1249,
      "num_input_tokens_seen": 39451608,
      "step": 24530
    },
    {
      "epoch": 1.110686072568415,
      "grad_norm": 1.1094833612442017,
      "learning_rate": 1.6284471929645338e-05,
      "loss": 0.1734,
      "num_input_tokens_seen": 39459928,
      "step": 24535
    },
    {
      "epoch": 1.110912424455058,
      "grad_norm": 0.6678844690322876,
      "learning_rate": 1.627527103668291e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 39467704,
      "step": 24540
    },
    {
      "epoch": 1.1111387763417009,
      "grad_norm": 0.6996273994445801,
      "learning_rate": 1.6266071489183327e-05,
      "loss": 0.152,
      "num_input_tokens_seen": 39475576,
      "step": 24545
    },
    {
      "epoch": 1.1113651282283439,
      "grad_norm": 0.5536805391311646,
      "learning_rate": 1.6256873288565283e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 39484120,
      "step": 24550
    },
    {
      "epoch": 1.1115914801149867,
      "grad_norm": 1.261462688446045,
      "learning_rate": 1.6247676436247245e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 39493336,
      "step": 24555
    },
    {
      "epoch": 1.1118178320016296,
      "grad_norm": 0.5446242690086365,
      "learning_rate": 1.6238480933647486e-05,
      "loss": 0.1258,
      "num_input_tokens_seen": 39501528,
      "step": 24560
    },
    {
      "epoch": 1.1120441838882726,
      "grad_norm": 0.8082068562507629,
      "learning_rate": 1.6229286782184083e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 39510040,
      "step": 24565
    },
    {
      "epoch": 1.1122705357749156,
      "grad_norm": 0.30778035521507263,
      "learning_rate": 1.622009398327487e-05,
      "loss": 0.1763,
      "num_input_tokens_seen": 39517976,
      "step": 24570
    },
    {
      "epoch": 1.1124968876615586,
      "grad_norm": 0.7670885920524597,
      "learning_rate": 1.6210902538337502e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 39526200,
      "step": 24575
    },
    {
      "epoch": 1.1127232395482016,
      "grad_norm": 0.4369361996650696,
      "learning_rate": 1.6201712448789413e-05,
      "loss": 0.1095,
      "num_input_tokens_seen": 39534136,
      "step": 24580
    },
    {
      "epoch": 1.1129495914348446,
      "grad_norm": 0.4841824471950531,
      "learning_rate": 1.6192523716047827e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 39542360,
      "step": 24585
    },
    {
      "epoch": 1.1131759433214876,
      "grad_norm": 1.4161500930786133,
      "learning_rate": 1.6183336341529776e-05,
      "loss": 0.1466,
      "num_input_tokens_seen": 39550808,
      "step": 24590
    },
    {
      "epoch": 1.1134022952081306,
      "grad_norm": 0.6956019401550293,
      "learning_rate": 1.6174150326652047e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 39558648,
      "step": 24595
    },
    {
      "epoch": 1.1136286470947736,
      "grad_norm": 0.5875191688537598,
      "learning_rate": 1.6164965672831256e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 39566872,
      "step": 24600
    },
    {
      "epoch": 1.1136286470947736,
      "eval_loss": 0.14514592289924622,
      "eval_runtime": 404.4172,
      "eval_samples_per_second": 97.105,
      "eval_steps_per_second": 24.277,
      "num_input_tokens_seen": 39566872,
      "step": 24600
    },
    {
      "epoch": 1.1138549989814166,
      "grad_norm": 1.0587222576141357,
      "learning_rate": 1.6155782381483784e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 39574552,
      "step": 24605
    },
    {
      "epoch": 1.1140813508680596,
      "grad_norm": 0.9292858839035034,
      "learning_rate": 1.6146600454025813e-05,
      "loss": 0.1683,
      "num_input_tokens_seen": 39582584,
      "step": 24610
    },
    {
      "epoch": 1.1143077027547024,
      "grad_norm": 0.6603311896324158,
      "learning_rate": 1.6137419891873317e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 39590264,
      "step": 24615
    },
    {
      "epoch": 1.1145340546413454,
      "grad_norm": 0.6382432579994202,
      "learning_rate": 1.6128240696442038e-05,
      "loss": 0.1786,
      "num_input_tokens_seen": 39598328,
      "step": 24620
    },
    {
      "epoch": 1.1147604065279884,
      "grad_norm": 0.4890870749950409,
      "learning_rate": 1.611906286914753e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 39606328,
      "step": 24625
    },
    {
      "epoch": 1.1149867584146314,
      "grad_norm": 0.53593510389328,
      "learning_rate": 1.6109886411405144e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 39614904,
      "step": 24630
    },
    {
      "epoch": 1.1152131103012743,
      "grad_norm": 0.6290415525436401,
      "learning_rate": 1.6100711324629985e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 39622456,
      "step": 24635
    },
    {
      "epoch": 1.1154394621879173,
      "grad_norm": 0.8070386648178101,
      "learning_rate": 1.609153761023698e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 39630712,
      "step": 24640
    },
    {
      "epoch": 1.1156658140745603,
      "grad_norm": 0.3150884509086609,
      "learning_rate": 1.608236526964083e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 39638488,
      "step": 24645
    },
    {
      "epoch": 1.1158921659612033,
      "grad_norm": 0.6639494895935059,
      "learning_rate": 1.607319430425601e-05,
      "loss": 0.1402,
      "num_input_tokens_seen": 39646424,
      "step": 24650
    },
    {
      "epoch": 1.1161185178478463,
      "grad_norm": 0.5298260450363159,
      "learning_rate": 1.606402471549682e-05,
      "loss": 0.1495,
      "num_input_tokens_seen": 39654360,
      "step": 24655
    },
    {
      "epoch": 1.1163448697344893,
      "grad_norm": 0.5763416886329651,
      "learning_rate": 1.6054856504777312e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 39662296,
      "step": 24660
    },
    {
      "epoch": 1.1165712216211323,
      "grad_norm": 0.5854518413543701,
      "learning_rate": 1.6045689673511334e-05,
      "loss": 0.1673,
      "num_input_tokens_seen": 39670520,
      "step": 24665
    },
    {
      "epoch": 1.116797573507775,
      "grad_norm": 1.1351807117462158,
      "learning_rate": 1.6036524223112548e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 39678520,
      "step": 24670
    },
    {
      "epoch": 1.117023925394418,
      "grad_norm": 0.7673963904380798,
      "learning_rate": 1.602736015499436e-05,
      "loss": 0.1347,
      "num_input_tokens_seen": 39686200,
      "step": 24675
    },
    {
      "epoch": 1.117250277281061,
      "grad_norm": 0.9113476872444153,
      "learning_rate": 1.601819747057e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 39694296,
      "step": 24680
    },
    {
      "epoch": 1.117476629167704,
      "grad_norm": 0.39597803354263306,
      "learning_rate": 1.6009036171252465e-05,
      "loss": 0.1014,
      "num_input_tokens_seen": 39702136,
      "step": 24685
    },
    {
      "epoch": 1.117702981054347,
      "grad_norm": 0.7672135829925537,
      "learning_rate": 1.599987625845453e-05,
      "loss": 0.1708,
      "num_input_tokens_seen": 39710616,
      "step": 24690
    },
    {
      "epoch": 1.11792933294099,
      "grad_norm": 0.5071096420288086,
      "learning_rate": 1.599071773358879e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 39719032,
      "step": 24695
    },
    {
      "epoch": 1.118155684827633,
      "grad_norm": 1.1168303489685059,
      "learning_rate": 1.598156059806758e-05,
      "loss": 0.1253,
      "num_input_tokens_seen": 39727672,
      "step": 24700
    },
    {
      "epoch": 1.118382036714276,
      "grad_norm": 0.534336268901825,
      "learning_rate": 1.5972404853303062e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 39735800,
      "step": 24705
    },
    {
      "epoch": 1.118608388600919,
      "grad_norm": 0.7392450571060181,
      "learning_rate": 1.5963250500707172e-05,
      "loss": 0.1404,
      "num_input_tokens_seen": 39744184,
      "step": 24710
    },
    {
      "epoch": 1.118834740487562,
      "grad_norm": 0.7632207870483398,
      "learning_rate": 1.5954097541691612e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 39752632,
      "step": 24715
    },
    {
      "epoch": 1.119061092374205,
      "grad_norm": 0.656223475933075,
      "learning_rate": 1.5944945977667884e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 39760888,
      "step": 24720
    },
    {
      "epoch": 1.119287444260848,
      "grad_norm": 0.5703248381614685,
      "learning_rate": 1.593579581004729e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 39769080,
      "step": 24725
    },
    {
      "epoch": 1.1195137961474908,
      "grad_norm": 0.7356541752815247,
      "learning_rate": 1.592664704024088e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 39777528,
      "step": 24730
    },
    {
      "epoch": 1.1197401480341338,
      "grad_norm": 0.4707704186439514,
      "learning_rate": 1.591749966965953e-05,
      "loss": 0.1292,
      "num_input_tokens_seen": 39785112,
      "step": 24735
    },
    {
      "epoch": 1.1199664999207768,
      "grad_norm": 1.0750621557235718,
      "learning_rate": 1.5908353699713856e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 39792888,
      "step": 24740
    },
    {
      "epoch": 1.1201928518074198,
      "grad_norm": 1.188683271408081,
      "learning_rate": 1.5899209131814298e-05,
      "loss": 0.1726,
      "num_input_tokens_seen": 39801528,
      "step": 24745
    },
    {
      "epoch": 1.1204192036940628,
      "grad_norm": 1.0887771844863892,
      "learning_rate": 1.5890065967371067e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 39809176,
      "step": 24750
    },
    {
      "epoch": 1.1206455555807058,
      "grad_norm": 0.555694043636322,
      "learning_rate": 1.5880924207794144e-05,
      "loss": 0.2216,
      "num_input_tokens_seen": 39817432,
      "step": 24755
    },
    {
      "epoch": 1.1208719074673488,
      "grad_norm": 0.7721044421195984,
      "learning_rate": 1.5871783854493298e-05,
      "loss": 0.1084,
      "num_input_tokens_seen": 39825528,
      "step": 24760
    },
    {
      "epoch": 1.1210982593539918,
      "grad_norm": 0.7172175049781799,
      "learning_rate": 1.5862644908878106e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 39833656,
      "step": 24765
    },
    {
      "epoch": 1.1213246112406348,
      "grad_norm": 0.5285066366195679,
      "learning_rate": 1.5853507372357885e-05,
      "loss": 0.1543,
      "num_input_tokens_seen": 39841432,
      "step": 24770
    },
    {
      "epoch": 1.1215509631272778,
      "grad_norm": 0.7265176773071289,
      "learning_rate": 1.5844371246341776e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 39848952,
      "step": 24775
    },
    {
      "epoch": 1.1217773150139205,
      "grad_norm": 0.6896268129348755,
      "learning_rate": 1.5835236532238674e-05,
      "loss": 0.1173,
      "num_input_tokens_seen": 39856760,
      "step": 24780
    },
    {
      "epoch": 1.1220036669005635,
      "grad_norm": 0.5939397215843201,
      "learning_rate": 1.582610323145727e-05,
      "loss": 0.1069,
      "num_input_tokens_seen": 39865016,
      "step": 24785
    },
    {
      "epoch": 1.1222300187872065,
      "grad_norm": 0.8864133954048157,
      "learning_rate": 1.5816971345406035e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 39872984,
      "step": 24790
    },
    {
      "epoch": 1.1224563706738495,
      "grad_norm": 1.2289581298828125,
      "learning_rate": 1.5807840875493225e-05,
      "loss": 0.1723,
      "num_input_tokens_seen": 39881272,
      "step": 24795
    },
    {
      "epoch": 1.1226827225604925,
      "grad_norm": 1.3041661977767944,
      "learning_rate": 1.5798711823126854e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 39889656,
      "step": 24800
    },
    {
      "epoch": 1.1226827225604925,
      "eval_loss": 0.14543239772319794,
      "eval_runtime": 404.6892,
      "eval_samples_per_second": 97.04,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 39889656,
      "step": 24800
    },
    {
      "epoch": 1.1229090744471355,
      "grad_norm": 0.7557793855667114,
      "learning_rate": 1.578958418971477e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 39897880,
      "step": 24805
    },
    {
      "epoch": 1.1231354263337785,
      "grad_norm": 0.4990924000740051,
      "learning_rate": 1.578045797666453e-05,
      "loss": 0.1314,
      "num_input_tokens_seen": 39906008,
      "step": 24810
    },
    {
      "epoch": 1.1233617782204215,
      "grad_norm": 0.5607994198799133,
      "learning_rate": 1.5771333185383548e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 39913176,
      "step": 24815
    },
    {
      "epoch": 1.1235881301070645,
      "grad_norm": 0.486843466758728,
      "learning_rate": 1.576220981727895e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 39920984,
      "step": 24820
    },
    {
      "epoch": 1.1238144819937075,
      "grad_norm": 1.0285011529922485,
      "learning_rate": 1.575308787375769e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 39929336,
      "step": 24825
    },
    {
      "epoch": 1.1240408338803505,
      "grad_norm": 0.8023092150688171,
      "learning_rate": 1.5743967356226492e-05,
      "loss": 0.1021,
      "num_input_tokens_seen": 39937016,
      "step": 24830
    },
    {
      "epoch": 1.1242671857669935,
      "grad_norm": 0.5915343165397644,
      "learning_rate": 1.5734848266091835e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 39945400,
      "step": 24835
    },
    {
      "epoch": 1.1244935376536362,
      "grad_norm": 0.6119633913040161,
      "learning_rate": 1.572573060476001e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 39953144,
      "step": 24840
    },
    {
      "epoch": 1.1247198895402792,
      "grad_norm": 0.4749859571456909,
      "learning_rate": 1.5716614373637085e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 39961848,
      "step": 24845
    },
    {
      "epoch": 1.1249462414269222,
      "grad_norm": 0.9474437832832336,
      "learning_rate": 1.570749957412887e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 39970936,
      "step": 24850
    },
    {
      "epoch": 1.1251725933135652,
      "grad_norm": 0.9512228965759277,
      "learning_rate": 1.5698386207641013e-05,
      "loss": 0.185,
      "num_input_tokens_seen": 39978872,
      "step": 24855
    },
    {
      "epoch": 1.1253989452002082,
      "grad_norm": 0.5002071261405945,
      "learning_rate": 1.5689274275578884e-05,
      "loss": 0.1537,
      "num_input_tokens_seen": 39986776,
      "step": 24860
    },
    {
      "epoch": 1.1256252970868512,
      "grad_norm": 0.8301410675048828,
      "learning_rate": 1.5680163779347667e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 39994776,
      "step": 24865
    },
    {
      "epoch": 1.1258516489734942,
      "grad_norm": 0.44363096356391907,
      "learning_rate": 1.5671054720352327e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 40002328,
      "step": 24870
    },
    {
      "epoch": 1.1260780008601372,
      "grad_norm": 0.8928701281547546,
      "learning_rate": 1.566194709999757e-05,
      "loss": 0.1034,
      "num_input_tokens_seen": 40010264,
      "step": 24875
    },
    {
      "epoch": 1.1263043527467802,
      "grad_norm": 0.6551102995872498,
      "learning_rate": 1.5652840919687933e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 40017944,
      "step": 24880
    },
    {
      "epoch": 1.1265307046334232,
      "grad_norm": 0.3885419964790344,
      "learning_rate": 1.5643736180827676e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 40025816,
      "step": 24885
    },
    {
      "epoch": 1.1267570565200662,
      "grad_norm": 0.5217960476875305,
      "learning_rate": 1.5634632884820878e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 40034296,
      "step": 24890
    },
    {
      "epoch": 1.126983408406709,
      "grad_norm": 0.5573713183403015,
      "learning_rate": 1.5625531033071395e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 40042584,
      "step": 24895
    },
    {
      "epoch": 1.127209760293352,
      "grad_norm": 0.38942062854766846,
      "learning_rate": 1.5616430626982828e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 40050968,
      "step": 24900
    },
    {
      "epoch": 1.127436112179995,
      "grad_norm": 0.8688512444496155,
      "learning_rate": 1.5607331667958575e-05,
      "loss": 0.1511,
      "num_input_tokens_seen": 40058808,
      "step": 24905
    },
    {
      "epoch": 1.127662464066638,
      "grad_norm": 1.4362608194351196,
      "learning_rate": 1.5598234157401824e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 40067352,
      "step": 24910
    },
    {
      "epoch": 1.127888815953281,
      "grad_norm": 0.39881354570388794,
      "learning_rate": 1.5589138096715503e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 40075576,
      "step": 24915
    },
    {
      "epoch": 1.128115167839924,
      "grad_norm": 0.3982393741607666,
      "learning_rate": 1.5580043487302365e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 40083832,
      "step": 24920
    },
    {
      "epoch": 1.128341519726567,
      "grad_norm": 0.8362053036689758,
      "learning_rate": 1.5570950330564888e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 40091448,
      "step": 24925
    },
    {
      "epoch": 1.12856787161321,
      "grad_norm": 0.48808398842811584,
      "learning_rate": 1.5561858627905367e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 40099512,
      "step": 24930
    },
    {
      "epoch": 1.128794223499853,
      "grad_norm": 0.2805968225002289,
      "learning_rate": 1.5552768380725857e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 40107064,
      "step": 24935
    },
    {
      "epoch": 1.129020575386496,
      "grad_norm": 1.3656233549118042,
      "learning_rate": 1.5543679590428183e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 40114616,
      "step": 24940
    },
    {
      "epoch": 1.129246927273139,
      "grad_norm": 0.622428297996521,
      "learning_rate": 1.5534592258413943e-05,
      "loss": 0.1322,
      "num_input_tokens_seen": 40122488,
      "step": 24945
    },
    {
      "epoch": 1.129473279159782,
      "grad_norm": 0.7347994446754456,
      "learning_rate": 1.5525506386084538e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 40129816,
      "step": 24950
    },
    {
      "epoch": 1.1296996310464247,
      "grad_norm": 0.48919183015823364,
      "learning_rate": 1.55164219748411e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 40138104,
      "step": 24955
    },
    {
      "epoch": 1.1299259829330677,
      "grad_norm": 0.6702699065208435,
      "learning_rate": 1.550733902608459e-05,
      "loss": 0.1465,
      "num_input_tokens_seen": 40146072,
      "step": 24960
    },
    {
      "epoch": 1.1301523348197107,
      "grad_norm": 0.3898298740386963,
      "learning_rate": 1.549825754121568e-05,
      "loss": 0.1331,
      "num_input_tokens_seen": 40154072,
      "step": 24965
    },
    {
      "epoch": 1.1303786867063537,
      "grad_norm": 0.44345414638519287,
      "learning_rate": 1.5489177521634864e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 40161912,
      "step": 24970
    },
    {
      "epoch": 1.1306050385929967,
      "grad_norm": 0.4669133424758911,
      "learning_rate": 1.5480098968742402e-05,
      "loss": 0.1217,
      "num_input_tokens_seen": 40169368,
      "step": 24975
    },
    {
      "epoch": 1.1308313904796397,
      "grad_norm": 0.9062824249267578,
      "learning_rate": 1.5471021883938304e-05,
      "loss": 0.1276,
      "num_input_tokens_seen": 40177176,
      "step": 24980
    },
    {
      "epoch": 1.1310577423662826,
      "grad_norm": 0.7722909450531006,
      "learning_rate": 1.546194626862238e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 40185272,
      "step": 24985
    },
    {
      "epoch": 1.1312840942529256,
      "grad_norm": 0.763814389705658,
      "learning_rate": 1.5452872124194216e-05,
      "loss": 0.1099,
      "num_input_tokens_seen": 40193656,
      "step": 24990
    },
    {
      "epoch": 1.1315104461395686,
      "grad_norm": 0.6887236833572388,
      "learning_rate": 1.5443799452053136e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 40201464,
      "step": 24995
    },
    {
      "epoch": 1.1317367980262116,
      "grad_norm": 0.449292927980423,
      "learning_rate": 1.543472825359828e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 40209528,
      "step": 25000
    },
    {
      "epoch": 1.1317367980262116,
      "eval_loss": 0.14522890746593475,
      "eval_runtime": 405.1985,
      "eval_samples_per_second": 96.918,
      "eval_steps_per_second": 24.23,
      "num_input_tokens_seen": 40209528,
      "step": 25000
    },
    {
      "epoch": 1.1319631499128544,
      "grad_norm": 0.5748594999313354,
      "learning_rate": 1.5425658530228522e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 40218040,
      "step": 25005
    },
    {
      "epoch": 1.1321895017994974,
      "grad_norm": 0.4491528570652008,
      "learning_rate": 1.5416590283342546e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 40225912,
      "step": 25010
    },
    {
      "epoch": 1.1324158536861404,
      "grad_norm": 0.7145559191703796,
      "learning_rate": 1.5407523514338783e-05,
      "loss": 0.1933,
      "num_input_tokens_seen": 40233880,
      "step": 25015
    },
    {
      "epoch": 1.1326422055727834,
      "grad_norm": 0.8044893145561218,
      "learning_rate": 1.539845822461543e-05,
      "loss": 0.1764,
      "num_input_tokens_seen": 40241496,
      "step": 25020
    },
    {
      "epoch": 1.1328685574594264,
      "grad_norm": 0.4408167600631714,
      "learning_rate": 1.538939441557048e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 40249784,
      "step": 25025
    },
    {
      "epoch": 1.1330949093460694,
      "grad_norm": 0.37174078822135925,
      "learning_rate": 1.5380332088601696e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 40257816,
      "step": 25030
    },
    {
      "epoch": 1.1333212612327124,
      "grad_norm": 0.5521445870399475,
      "learning_rate": 1.537127124510658e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 40266040,
      "step": 25035
    },
    {
      "epoch": 1.1335476131193554,
      "grad_norm": 0.5217809081077576,
      "learning_rate": 1.5362211886482457e-05,
      "loss": 0.1428,
      "num_input_tokens_seen": 40274584,
      "step": 25040
    },
    {
      "epoch": 1.1337739650059984,
      "grad_norm": 0.8645196557044983,
      "learning_rate": 1.5353154014126363e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 40282296,
      "step": 25045
    },
    {
      "epoch": 1.1340003168926414,
      "grad_norm": 0.8866434097290039,
      "learning_rate": 1.534409762943515e-05,
      "loss": 0.1165,
      "num_input_tokens_seen": 40290520,
      "step": 25050
    },
    {
      "epoch": 1.1342266687792844,
      "grad_norm": 0.5180999040603638,
      "learning_rate": 1.5335042733805438e-05,
      "loss": 0.1515,
      "num_input_tokens_seen": 40298840,
      "step": 25055
    },
    {
      "epoch": 1.1344530206659273,
      "grad_norm": 0.5181557536125183,
      "learning_rate": 1.532598932863358e-05,
      "loss": 0.1035,
      "num_input_tokens_seen": 40307000,
      "step": 25060
    },
    {
      "epoch": 1.1346793725525703,
      "grad_norm": 0.9144002199172974,
      "learning_rate": 1.531693741531574e-05,
      "loss": 0.1646,
      "num_input_tokens_seen": 40314904,
      "step": 25065
    },
    {
      "epoch": 1.1349057244392131,
      "grad_norm": 0.7556180953979492,
      "learning_rate": 1.5307886995247844e-05,
      "loss": 0.1462,
      "num_input_tokens_seen": 40323320,
      "step": 25070
    },
    {
      "epoch": 1.135132076325856,
      "grad_norm": 0.514242947101593,
      "learning_rate": 1.529883806982557e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 40331736,
      "step": 25075
    },
    {
      "epoch": 1.135358428212499,
      "grad_norm": 0.8234224319458008,
      "learning_rate": 1.5289790640444376e-05,
      "loss": 0.1414,
      "num_input_tokens_seen": 40340824,
      "step": 25080
    },
    {
      "epoch": 1.135584780099142,
      "grad_norm": 0.2986201345920563,
      "learning_rate": 1.5280744708499494e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 40349016,
      "step": 25085
    },
    {
      "epoch": 1.135811131985785,
      "grad_norm": 1.5807223320007324,
      "learning_rate": 1.527170027538591e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 40357432,
      "step": 25090
    },
    {
      "epoch": 1.136037483872428,
      "grad_norm": 0.5453716516494751,
      "learning_rate": 1.5262657342498407e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 40364952,
      "step": 25095
    },
    {
      "epoch": 1.136263835759071,
      "grad_norm": 0.3357272446155548,
      "learning_rate": 1.52536159112315e-05,
      "loss": 0.1096,
      "num_input_tokens_seen": 40372568,
      "step": 25100
    },
    {
      "epoch": 1.136490187645714,
      "grad_norm": 0.9650960564613342,
      "learning_rate": 1.5244575982979497e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 40380824,
      "step": 25105
    },
    {
      "epoch": 1.136716539532357,
      "grad_norm": 0.808008074760437,
      "learning_rate": 1.5235537559136487e-05,
      "loss": 0.1439,
      "num_input_tokens_seen": 40389976,
      "step": 25110
    },
    {
      "epoch": 1.136942891419,
      "grad_norm": 0.8700704574584961,
      "learning_rate": 1.5226500641096286e-05,
      "loss": 0.155,
      "num_input_tokens_seen": 40398072,
      "step": 25115
    },
    {
      "epoch": 1.1371692433056428,
      "grad_norm": 0.9531251788139343,
      "learning_rate": 1.5217465230252509e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 40405944,
      "step": 25120
    },
    {
      "epoch": 1.1373955951922858,
      "grad_norm": 0.7561204433441162,
      "learning_rate": 1.5208431327998523e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 40414680,
      "step": 25125
    },
    {
      "epoch": 1.1376219470789288,
      "grad_norm": 0.5903701186180115,
      "learning_rate": 1.5199398935727477e-05,
      "loss": 0.137,
      "num_input_tokens_seen": 40422648,
      "step": 25130
    },
    {
      "epoch": 1.1378482989655718,
      "grad_norm": 1.0827728509902954,
      "learning_rate": 1.5190368054832282e-05,
      "loss": 0.1721,
      "num_input_tokens_seen": 40430808,
      "step": 25135
    },
    {
      "epoch": 1.1380746508522148,
      "grad_norm": 2.0430915355682373,
      "learning_rate": 1.5181338686705601e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 40438616,
      "step": 25140
    },
    {
      "epoch": 1.1383010027388578,
      "grad_norm": 0.5551556348800659,
      "learning_rate": 1.5172310832739889e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 40446584,
      "step": 25145
    },
    {
      "epoch": 1.1385273546255008,
      "grad_norm": 0.8441834449768066,
      "learning_rate": 1.5163284494327346e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 40454744,
      "step": 25150
    },
    {
      "epoch": 1.1387537065121438,
      "grad_norm": 0.6465896964073181,
      "learning_rate": 1.5154259672859952e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 40462584,
      "step": 25155
    },
    {
      "epoch": 1.1389800583987868,
      "grad_norm": 1.2157291173934937,
      "learning_rate": 1.5145236369729452e-05,
      "loss": 0.1406,
      "num_input_tokens_seen": 40470840,
      "step": 25160
    },
    {
      "epoch": 1.1392064102854298,
      "grad_norm": 0.4574684798717499,
      "learning_rate": 1.5136214586327335e-05,
      "loss": 0.169,
      "num_input_tokens_seen": 40478680,
      "step": 25165
    },
    {
      "epoch": 1.1394327621720728,
      "grad_norm": 0.4771249294281006,
      "learning_rate": 1.5127194324044885e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 40486488,
      "step": 25170
    },
    {
      "epoch": 1.1396591140587158,
      "grad_norm": 0.5881194472312927,
      "learning_rate": 1.5118175584273148e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 40494456,
      "step": 25175
    },
    {
      "epoch": 1.1398854659453588,
      "grad_norm": 1.2762181758880615,
      "learning_rate": 1.5109158368402909e-05,
      "loss": 0.1312,
      "num_input_tokens_seen": 40502296,
      "step": 25180
    },
    {
      "epoch": 1.1401118178320016,
      "grad_norm": 0.7067702412605286,
      "learning_rate": 1.5100142677824753e-05,
      "loss": 0.1372,
      "num_input_tokens_seen": 40510296,
      "step": 25185
    },
    {
      "epoch": 1.1403381697186445,
      "grad_norm": 0.9394193887710571,
      "learning_rate": 1.509112851392901e-05,
      "loss": 0.1628,
      "num_input_tokens_seen": 40518872,
      "step": 25190
    },
    {
      "epoch": 1.1405645216052875,
      "grad_norm": 0.40372008085250854,
      "learning_rate": 1.5082115878105763e-05,
      "loss": 0.1798,
      "num_input_tokens_seen": 40526936,
      "step": 25195
    },
    {
      "epoch": 1.1407908734919305,
      "grad_norm": 0.5455406904220581,
      "learning_rate": 1.5073104771744892e-05,
      "loss": 0.1206,
      "num_input_tokens_seen": 40535352,
      "step": 25200
    },
    {
      "epoch": 1.1407908734919305,
      "eval_loss": 0.1451113373041153,
      "eval_runtime": 404.6707,
      "eval_samples_per_second": 97.044,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 40535352,
      "step": 25200
    },
    {
      "epoch": 1.1410172253785735,
      "grad_norm": 0.9355839490890503,
      "learning_rate": 1.5064095196236006e-05,
      "loss": 0.1754,
      "num_input_tokens_seen": 40543960,
      "step": 25205
    },
    {
      "epoch": 1.1412435772652165,
      "grad_norm": 0.6598964929580688,
      "learning_rate": 1.50550871529685e-05,
      "loss": 0.146,
      "num_input_tokens_seen": 40551096,
      "step": 25210
    },
    {
      "epoch": 1.1414699291518595,
      "grad_norm": 0.3817162811756134,
      "learning_rate": 1.5046080643331546e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 40558904,
      "step": 25215
    },
    {
      "epoch": 1.1416962810385025,
      "grad_norm": 0.3870674967765808,
      "learning_rate": 1.5037075668714028e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 40566744,
      "step": 25220
    },
    {
      "epoch": 1.1419226329251455,
      "grad_norm": 0.7729936838150024,
      "learning_rate": 1.5028072230504656e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 40574200,
      "step": 25225
    },
    {
      "epoch": 1.1421489848117883,
      "grad_norm": 1.0801554918289185,
      "learning_rate": 1.5019070330091861e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 40582424,
      "step": 25230
    },
    {
      "epoch": 1.1423753366984313,
      "grad_norm": 0.9515220522880554,
      "learning_rate": 1.5010069968863843e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 40590456,
      "step": 25235
    },
    {
      "epoch": 1.1426016885850743,
      "grad_norm": 0.9581525921821594,
      "learning_rate": 1.5001071148208584e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 40598520,
      "step": 25240
    },
    {
      "epoch": 1.1428280404717173,
      "grad_norm": 0.37454450130462646,
      "learning_rate": 1.49920738695138e-05,
      "loss": 0.126,
      "num_input_tokens_seen": 40606328,
      "step": 25245
    },
    {
      "epoch": 1.1430543923583603,
      "grad_norm": 0.4633215069770813,
      "learning_rate": 1.4983078134166995e-05,
      "loss": 0.102,
      "num_input_tokens_seen": 40614456,
      "step": 25250
    },
    {
      "epoch": 1.1432807442450033,
      "grad_norm": 0.5068860054016113,
      "learning_rate": 1.4974083943555428e-05,
      "loss": 0.1529,
      "num_input_tokens_seen": 40621912,
      "step": 25255
    },
    {
      "epoch": 1.1435070961316463,
      "grad_norm": 0.5688866376876831,
      "learning_rate": 1.496509129906611e-05,
      "loss": 0.1626,
      "num_input_tokens_seen": 40629464,
      "step": 25260
    },
    {
      "epoch": 1.1437334480182892,
      "grad_norm": 0.29967233538627625,
      "learning_rate": 1.4956100202085809e-05,
      "loss": 0.1007,
      "num_input_tokens_seen": 40637656,
      "step": 25265
    },
    {
      "epoch": 1.1439597999049322,
      "grad_norm": 0.7906714677810669,
      "learning_rate": 1.4947110654001093e-05,
      "loss": 0.1831,
      "num_input_tokens_seen": 40645784,
      "step": 25270
    },
    {
      "epoch": 1.1441861517915752,
      "grad_norm": 0.9239667654037476,
      "learning_rate": 1.4938122656198234e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 40653880,
      "step": 25275
    },
    {
      "epoch": 1.1444125036782182,
      "grad_norm": 0.828656017780304,
      "learning_rate": 1.4929136210063316e-05,
      "loss": 0.1237,
      "num_input_tokens_seen": 40661976,
      "step": 25280
    },
    {
      "epoch": 1.1446388555648612,
      "grad_norm": 0.9699068069458008,
      "learning_rate": 1.4920151316982146e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 40669976,
      "step": 25285
    },
    {
      "epoch": 1.1448652074515042,
      "grad_norm": 0.6625978946685791,
      "learning_rate": 1.4911167978340312e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 40678648,
      "step": 25290
    },
    {
      "epoch": 1.145091559338147,
      "grad_norm": 0.6834924817085266,
      "learning_rate": 1.4902186195523166e-05,
      "loss": 0.122,
      "num_input_tokens_seen": 40687224,
      "step": 25295
    },
    {
      "epoch": 1.14531791122479,
      "grad_norm": 0.7134552001953125,
      "learning_rate": 1.4893205969915805e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 40696632,
      "step": 25300
    },
    {
      "epoch": 1.145544263111433,
      "grad_norm": 0.868184506893158,
      "learning_rate": 1.4884227302903086e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 40704696,
      "step": 25305
    },
    {
      "epoch": 1.145770614998076,
      "grad_norm": 0.4344744384288788,
      "learning_rate": 1.4875250195869653e-05,
      "loss": 0.1158,
      "num_input_tokens_seen": 40713496,
      "step": 25310
    },
    {
      "epoch": 1.145996966884719,
      "grad_norm": 1.2442755699157715,
      "learning_rate": 1.4866274650199862e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 40720824,
      "step": 25315
    },
    {
      "epoch": 1.146223318771362,
      "grad_norm": 0.4736168384552002,
      "learning_rate": 1.485730066727788e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 40728728,
      "step": 25320
    },
    {
      "epoch": 1.146449670658005,
      "grad_norm": 0.47621747851371765,
      "learning_rate": 1.4848328248487586e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 40736472,
      "step": 25325
    },
    {
      "epoch": 1.146676022544648,
      "grad_norm": 0.8358595967292786,
      "learning_rate": 1.4839357395212656e-05,
      "loss": 0.1759,
      "num_input_tokens_seen": 40745368,
      "step": 25330
    },
    {
      "epoch": 1.146902374431291,
      "grad_norm": 0.5950184464454651,
      "learning_rate": 1.4830388108836502e-05,
      "loss": 0.1635,
      "num_input_tokens_seen": 40753368,
      "step": 25335
    },
    {
      "epoch": 1.147128726317934,
      "grad_norm": 0.8094961047172546,
      "learning_rate": 1.4821420390742299e-05,
      "loss": 0.1585,
      "num_input_tokens_seen": 40761080,
      "step": 25340
    },
    {
      "epoch": 1.1473550782045767,
      "grad_norm": 0.5784975290298462,
      "learning_rate": 1.4812454242312979e-05,
      "loss": 0.1171,
      "num_input_tokens_seen": 40769240,
      "step": 25345
    },
    {
      "epoch": 1.1475814300912197,
      "grad_norm": 0.6139585971832275,
      "learning_rate": 1.4803489664931253e-05,
      "loss": 0.1267,
      "num_input_tokens_seen": 40776888,
      "step": 25350
    },
    {
      "epoch": 1.1478077819778627,
      "grad_norm": 0.48939159512519836,
      "learning_rate": 1.4794526659979544e-05,
      "loss": 0.1553,
      "num_input_tokens_seen": 40784856,
      "step": 25355
    },
    {
      "epoch": 1.1480341338645057,
      "grad_norm": 0.6433011293411255,
      "learning_rate": 1.4785565228840086e-05,
      "loss": 0.1152,
      "num_input_tokens_seen": 40793464,
      "step": 25360
    },
    {
      "epoch": 1.1482604857511487,
      "grad_norm": 0.8585342764854431,
      "learning_rate": 1.4776605372894819e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 40801368,
      "step": 25365
    },
    {
      "epoch": 1.1484868376377917,
      "grad_norm": 1.247771978378296,
      "learning_rate": 1.4767647093525488e-05,
      "loss": 0.1422,
      "num_input_tokens_seen": 40809176,
      "step": 25370
    },
    {
      "epoch": 1.1487131895244347,
      "grad_norm": 0.43942296504974365,
      "learning_rate": 1.4758690392113566e-05,
      "loss": 0.1199,
      "num_input_tokens_seen": 40816856,
      "step": 25375
    },
    {
      "epoch": 1.1489395414110777,
      "grad_norm": 0.5496721267700195,
      "learning_rate": 1.4749735270040276e-05,
      "loss": 0.1447,
      "num_input_tokens_seen": 40825272,
      "step": 25380
    },
    {
      "epoch": 1.1491658932977207,
      "grad_norm": 0.935126006603241,
      "learning_rate": 1.4740781728686623e-05,
      "loss": 0.1262,
      "num_input_tokens_seen": 40833784,
      "step": 25385
    },
    {
      "epoch": 1.1493922451843637,
      "grad_norm": 0.8440605998039246,
      "learning_rate": 1.4731829769433358e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 40842104,
      "step": 25390
    },
    {
      "epoch": 1.1496185970710067,
      "grad_norm": 0.709210216999054,
      "learning_rate": 1.4722879393660976e-05,
      "loss": 0.1284,
      "num_input_tokens_seen": 40850232,
      "step": 25395
    },
    {
      "epoch": 1.1498449489576497,
      "grad_norm": 0.32114094495773315,
      "learning_rate": 1.4713930602749748e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 40858264,
      "step": 25400
    },
    {
      "epoch": 1.1498449489576497,
      "eval_loss": 0.14480803906917572,
      "eval_runtime": 405.2643,
      "eval_samples_per_second": 96.902,
      "eval_steps_per_second": 24.226,
      "num_input_tokens_seen": 40858264,
      "step": 25400
    },
    {
      "epoch": 1.1500713008442927,
      "grad_norm": 0.4663294553756714,
      "learning_rate": 1.470498339807968e-05,
      "loss": 0.1113,
      "num_input_tokens_seen": 40866296,
      "step": 25405
    },
    {
      "epoch": 1.1502976527309354,
      "grad_norm": 0.8924300074577332,
      "learning_rate": 1.4696037781030542e-05,
      "loss": 0.1085,
      "num_input_tokens_seen": 40874456,
      "step": 25410
    },
    {
      "epoch": 1.1505240046175784,
      "grad_norm": 0.6614165306091309,
      "learning_rate": 1.4687093752981876e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 40882424,
      "step": 25415
    },
    {
      "epoch": 1.1507503565042214,
      "grad_norm": 0.7117006182670593,
      "learning_rate": 1.4678151315312943e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 40890520,
      "step": 25420
    },
    {
      "epoch": 1.1509767083908644,
      "grad_norm": 0.7098137736320496,
      "learning_rate": 1.4669210469402789e-05,
      "loss": 0.1202,
      "num_input_tokens_seen": 40898712,
      "step": 25425
    },
    {
      "epoch": 1.1512030602775074,
      "grad_norm": 1.0813958644866943,
      "learning_rate": 1.4660271216630218e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 40906776,
      "step": 25430
    },
    {
      "epoch": 1.1514294121641504,
      "grad_norm": 0.7334558367729187,
      "learning_rate": 1.4651333558373748e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 40914552,
      "step": 25435
    },
    {
      "epoch": 1.1516557640507934,
      "grad_norm": 0.8327096104621887,
      "learning_rate": 1.4642397496011707e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 40922552,
      "step": 25440
    },
    {
      "epoch": 1.1518821159374364,
      "grad_norm": 0.3713620901107788,
      "learning_rate": 1.4633463030922129e-05,
      "loss": 0.1229,
      "num_input_tokens_seen": 40930328,
      "step": 25445
    },
    {
      "epoch": 1.1521084678240794,
      "grad_norm": 0.8720389008522034,
      "learning_rate": 1.462453016448282e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 40938040,
      "step": 25450
    },
    {
      "epoch": 1.1523348197107224,
      "grad_norm": 0.401920884847641,
      "learning_rate": 1.4615598898071354e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 40947000,
      "step": 25455
    },
    {
      "epoch": 1.1525611715973652,
      "grad_norm": 0.5553339719772339,
      "learning_rate": 1.4606669233065026e-05,
      "loss": 0.1391,
      "num_input_tokens_seen": 40954552,
      "step": 25460
    },
    {
      "epoch": 1.1527875234840081,
      "grad_norm": 0.6439455151557922,
      "learning_rate": 1.4597741170840914e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 40962744,
      "step": 25465
    },
    {
      "epoch": 1.1530138753706511,
      "grad_norm": 0.472494900226593,
      "learning_rate": 1.4588814712775853e-05,
      "loss": 0.1232,
      "num_input_tokens_seen": 40970552,
      "step": 25470
    },
    {
      "epoch": 1.1532402272572941,
      "grad_norm": 0.7346789836883545,
      "learning_rate": 1.4579889860246382e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 40978712,
      "step": 25475
    },
    {
      "epoch": 1.1534665791439371,
      "grad_norm": 1.2272679805755615,
      "learning_rate": 1.457096661462885e-05,
      "loss": 0.1434,
      "num_input_tokens_seen": 40986168,
      "step": 25480
    },
    {
      "epoch": 1.1536929310305801,
      "grad_norm": 0.8998847007751465,
      "learning_rate": 1.4562044977299322e-05,
      "loss": 0.1767,
      "num_input_tokens_seen": 40993624,
      "step": 25485
    },
    {
      "epoch": 1.1539192829172231,
      "grad_norm": 0.3826526701450348,
      "learning_rate": 1.4553124949633623e-05,
      "loss": 0.177,
      "num_input_tokens_seen": 41001752,
      "step": 25490
    },
    {
      "epoch": 1.1541456348038661,
      "grad_norm": 0.4817856252193451,
      "learning_rate": 1.4544206533007354e-05,
      "loss": 0.1917,
      "num_input_tokens_seen": 41009784,
      "step": 25495
    },
    {
      "epoch": 1.154371986690509,
      "grad_norm": 0.9378728866577148,
      "learning_rate": 1.4535289728795821e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 41018232,
      "step": 25500
    },
    {
      "epoch": 1.154598338577152,
      "grad_norm": 0.3337802588939667,
      "learning_rate": 1.4526374538374132e-05,
      "loss": 0.1317,
      "num_input_tokens_seen": 41026904,
      "step": 25505
    },
    {
      "epoch": 1.154824690463795,
      "grad_norm": 0.5136017799377441,
      "learning_rate": 1.4517460963117097e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 41035064,
      "step": 25510
    },
    {
      "epoch": 1.155051042350438,
      "grad_norm": 0.4186899662017822,
      "learning_rate": 1.4508549004399314e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 41043096,
      "step": 25515
    },
    {
      "epoch": 1.155277394237081,
      "grad_norm": 0.3525925874710083,
      "learning_rate": 1.449963866359513e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 41051448,
      "step": 25520
    },
    {
      "epoch": 1.1555037461237239,
      "grad_norm": 0.6894173622131348,
      "learning_rate": 1.4490729942078607e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 41059768,
      "step": 25525
    },
    {
      "epoch": 1.1557300980103669,
      "grad_norm": 0.7104902863502502,
      "learning_rate": 1.4481822841223608e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 41068280,
      "step": 25530
    },
    {
      "epoch": 1.1559564498970099,
      "grad_norm": 0.9324904680252075,
      "learning_rate": 1.4472917362403704e-05,
      "loss": 0.1606,
      "num_input_tokens_seen": 41076280,
      "step": 25535
    },
    {
      "epoch": 1.1561828017836528,
      "grad_norm": 0.5566760897636414,
      "learning_rate": 1.4464013506992224e-05,
      "loss": 0.1986,
      "num_input_tokens_seen": 41084280,
      "step": 25540
    },
    {
      "epoch": 1.1564091536702958,
      "grad_norm": 0.7383235096931458,
      "learning_rate": 1.4455111276362277e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 41092248,
      "step": 25545
    },
    {
      "epoch": 1.1566355055569388,
      "grad_norm": 0.8105968236923218,
      "learning_rate": 1.4446210671886676e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 41100504,
      "step": 25550
    },
    {
      "epoch": 1.1568618574435818,
      "grad_norm": 0.9538835883140564,
      "learning_rate": 1.4437311694938015e-05,
      "loss": 0.1516,
      "num_input_tokens_seen": 41108312,
      "step": 25555
    },
    {
      "epoch": 1.1570882093302248,
      "grad_norm": 0.3147117793560028,
      "learning_rate": 1.442841434688864e-05,
      "loss": 0.1679,
      "num_input_tokens_seen": 41116280,
      "step": 25560
    },
    {
      "epoch": 1.1573145612168678,
      "grad_norm": 1.1339765787124634,
      "learning_rate": 1.4419518629110615e-05,
      "loss": 0.173,
      "num_input_tokens_seen": 41124120,
      "step": 25565
    },
    {
      "epoch": 1.1575409131035106,
      "grad_norm": 0.33489087224006653,
      "learning_rate": 1.4410624542975778e-05,
      "loss": 0.11,
      "num_input_tokens_seen": 41132440,
      "step": 25570
    },
    {
      "epoch": 1.1577672649901536,
      "grad_norm": 0.49095386266708374,
      "learning_rate": 1.4401732089855724e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 41140440,
      "step": 25575
    },
    {
      "epoch": 1.1579936168767966,
      "grad_norm": 0.9031651616096497,
      "learning_rate": 1.4392841271121754e-05,
      "loss": 0.1625,
      "num_input_tokens_seen": 41149272,
      "step": 25580
    },
    {
      "epoch": 1.1582199687634396,
      "grad_norm": 0.40613245964050293,
      "learning_rate": 1.438395208814497e-05,
      "loss": 0.1444,
      "num_input_tokens_seen": 41157464,
      "step": 25585
    },
    {
      "epoch": 1.1584463206500826,
      "grad_norm": 0.5252603888511658,
      "learning_rate": 1.4375064542296174e-05,
      "loss": 0.1458,
      "num_input_tokens_seen": 41165720,
      "step": 25590
    },
    {
      "epoch": 1.1586726725367256,
      "grad_norm": 1.2313376665115356,
      "learning_rate": 1.4366178634945946e-05,
      "loss": 0.1357,
      "num_input_tokens_seen": 41173400,
      "step": 25595
    },
    {
      "epoch": 1.1588990244233686,
      "grad_norm": 1.6613550186157227,
      "learning_rate": 1.4357294367464616e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 41182296,
      "step": 25600
    },
    {
      "epoch": 1.1588990244233686,
      "eval_loss": 0.14522553980350494,
      "eval_runtime": 404.9495,
      "eval_samples_per_second": 96.978,
      "eval_steps_per_second": 24.245,
      "num_input_tokens_seen": 41182296,
      "step": 25600
    },
    {
      "epoch": 1.1591253763100116,
      "grad_norm": 0.6796004176139832,
      "learning_rate": 1.434841174122224e-05,
      "loss": 0.1063,
      "num_input_tokens_seen": 41189784,
      "step": 25605
    },
    {
      "epoch": 1.1593517281966546,
      "grad_norm": 0.5601744651794434,
      "learning_rate": 1.4339530757588615e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 41197752,
      "step": 25610
    },
    {
      "epoch": 1.1595780800832975,
      "grad_norm": 0.6138657331466675,
      "learning_rate": 1.433065141793333e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 41205464,
      "step": 25615
    },
    {
      "epoch": 1.1598044319699405,
      "grad_norm": 0.725212812423706,
      "learning_rate": 1.4321773723625665e-05,
      "loss": 0.178,
      "num_input_tokens_seen": 41214456,
      "step": 25620
    },
    {
      "epoch": 1.1600307838565835,
      "grad_norm": 0.548739492893219,
      "learning_rate": 1.4312897676034693e-05,
      "loss": 0.1937,
      "num_input_tokens_seen": 41222392,
      "step": 25625
    },
    {
      "epoch": 1.1602571357432265,
      "grad_norm": 0.7573323249816895,
      "learning_rate": 1.4304023276529188e-05,
      "loss": 0.166,
      "num_input_tokens_seen": 41229784,
      "step": 25630
    },
    {
      "epoch": 1.1604834876298693,
      "grad_norm": 0.6696792244911194,
      "learning_rate": 1.4295150526477712e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 41238328,
      "step": 25635
    },
    {
      "epoch": 1.1607098395165123,
      "grad_norm": 0.9318974018096924,
      "learning_rate": 1.4286279427248562e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 41245624,
      "step": 25640
    },
    {
      "epoch": 1.1609361914031553,
      "grad_norm": 0.2746313214302063,
      "learning_rate": 1.4277409980209747e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 41253144,
      "step": 25645
    },
    {
      "epoch": 1.1611625432897983,
      "grad_norm": 0.5858741998672485,
      "learning_rate": 1.4268542186729061e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 41261208,
      "step": 25650
    },
    {
      "epoch": 1.1613888951764413,
      "grad_norm": 0.3389820456504822,
      "learning_rate": 1.4259676048174043e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 41268984,
      "step": 25655
    },
    {
      "epoch": 1.1616152470630843,
      "grad_norm": 0.7175173163414001,
      "learning_rate": 1.4250811565911937e-05,
      "loss": 0.1782,
      "num_input_tokens_seen": 41276632,
      "step": 25660
    },
    {
      "epoch": 1.1618415989497273,
      "grad_norm": 0.5826711058616638,
      "learning_rate": 1.4241948741309782e-05,
      "loss": 0.1195,
      "num_input_tokens_seen": 41284312,
      "step": 25665
    },
    {
      "epoch": 1.1620679508363703,
      "grad_norm": 0.38116976618766785,
      "learning_rate": 1.4233087575734317e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 41292504,
      "step": 25670
    },
    {
      "epoch": 1.1622943027230133,
      "grad_norm": 0.5217793583869934,
      "learning_rate": 1.422422807055206e-05,
      "loss": 0.1275,
      "num_input_tokens_seen": 41300728,
      "step": 25675
    },
    {
      "epoch": 1.1625206546096563,
      "grad_norm": 0.31903067231178284,
      "learning_rate": 1.4215370227129243e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 41308600,
      "step": 25680
    },
    {
      "epoch": 1.162747006496299,
      "grad_norm": 0.5794060826301575,
      "learning_rate": 1.4206514046831876e-05,
      "loss": 0.1527,
      "num_input_tokens_seen": 41316088,
      "step": 25685
    },
    {
      "epoch": 1.162973358382942,
      "grad_norm": 0.36037522554397583,
      "learning_rate": 1.419765953102567e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 41324248,
      "step": 25690
    },
    {
      "epoch": 1.163199710269585,
      "grad_norm": 0.3946351706981659,
      "learning_rate": 1.4188806681076125e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 41332568,
      "step": 25695
    },
    {
      "epoch": 1.163426062156228,
      "grad_norm": 0.6944392323493958,
      "learning_rate": 1.4179955498348443e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 41340472,
      "step": 25700
    },
    {
      "epoch": 1.163652414042871,
      "grad_norm": 0.5809990763664246,
      "learning_rate": 1.4171105984207605e-05,
      "loss": 0.1166,
      "num_input_tokens_seen": 41348312,
      "step": 25705
    },
    {
      "epoch": 1.163878765929514,
      "grad_norm": 1.0188910961151123,
      "learning_rate": 1.4162258140018304e-05,
      "loss": 0.1486,
      "num_input_tokens_seen": 41356344,
      "step": 25710
    },
    {
      "epoch": 1.164105117816157,
      "grad_norm": 0.3710465729236603,
      "learning_rate": 1.4153411967144986e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 41364344,
      "step": 25715
    },
    {
      "epoch": 1.1643314697028,
      "grad_norm": 0.7625702619552612,
      "learning_rate": 1.4144567466951864e-05,
      "loss": 0.1309,
      "num_input_tokens_seen": 41372408,
      "step": 25720
    },
    {
      "epoch": 1.164557821589443,
      "grad_norm": 0.5521788597106934,
      "learning_rate": 1.4135724640802844e-05,
      "loss": 0.1666,
      "num_input_tokens_seen": 41380696,
      "step": 25725
    },
    {
      "epoch": 1.164784173476086,
      "grad_norm": 0.3479423522949219,
      "learning_rate": 1.4126883490061615e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 41389432,
      "step": 25730
    },
    {
      "epoch": 1.165010525362729,
      "grad_norm": 0.48810386657714844,
      "learning_rate": 1.4118044016091603e-05,
      "loss": 0.1242,
      "num_input_tokens_seen": 41397976,
      "step": 25735
    },
    {
      "epoch": 1.165236877249372,
      "grad_norm": 0.3841364085674286,
      "learning_rate": 1.410920622025594e-05,
      "loss": 0.1494,
      "num_input_tokens_seen": 41405944,
      "step": 25740
    },
    {
      "epoch": 1.165463229136015,
      "grad_norm": 0.37078094482421875,
      "learning_rate": 1.4100370103917554e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 41413944,
      "step": 25745
    },
    {
      "epoch": 1.1656895810226577,
      "grad_norm": 0.9405906796455383,
      "learning_rate": 1.409153566843907e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 41422136,
      "step": 25750
    },
    {
      "epoch": 1.1659159329093007,
      "grad_norm": 1.0290343761444092,
      "learning_rate": 1.408270291518286e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 41429752,
      "step": 25755
    },
    {
      "epoch": 1.1661422847959437,
      "grad_norm": 0.6325608491897583,
      "learning_rate": 1.407387184551107e-05,
      "loss": 0.1541,
      "num_input_tokens_seen": 41437656,
      "step": 25760
    },
    {
      "epoch": 1.1663686366825867,
      "grad_norm": 0.4255514144897461,
      "learning_rate": 1.4065042460785532e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 41445688,
      "step": 25765
    },
    {
      "epoch": 1.1665949885692297,
      "grad_norm": 0.4746854305267334,
      "learning_rate": 1.405621476236787e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 41453976,
      "step": 25770
    },
    {
      "epoch": 1.1668213404558727,
      "grad_norm": 0.49082526564598083,
      "learning_rate": 1.4047388751619423e-05,
      "loss": 0.1171,
      "num_input_tokens_seen": 41462232,
      "step": 25775
    },
    {
      "epoch": 1.1670476923425157,
      "grad_norm": 0.4687267243862152,
      "learning_rate": 1.4038564429901264e-05,
      "loss": 0.1212,
      "num_input_tokens_seen": 41470552,
      "step": 25780
    },
    {
      "epoch": 1.1672740442291587,
      "grad_norm": 0.8208550214767456,
      "learning_rate": 1.4029741798574227e-05,
      "loss": 0.1133,
      "num_input_tokens_seen": 41478328,
      "step": 25785
    },
    {
      "epoch": 1.1675003961158017,
      "grad_norm": 0.9686360359191895,
      "learning_rate": 1.402092085899886e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 41486328,
      "step": 25790
    },
    {
      "epoch": 1.1677267480024447,
      "grad_norm": 0.49944841861724854,
      "learning_rate": 1.4012101612535464e-05,
      "loss": 0.1479,
      "num_input_tokens_seen": 41494392,
      "step": 25795
    },
    {
      "epoch": 1.1679530998890875,
      "grad_norm": 0.7684727311134338,
      "learning_rate": 1.4003284060544092e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 41502296,
      "step": 25800
    },
    {
      "epoch": 1.1679530998890875,
      "eval_loss": 0.14484603703022003,
      "eval_runtime": 404.9459,
      "eval_samples_per_second": 96.978,
      "eval_steps_per_second": 24.245,
      "num_input_tokens_seen": 41502296,
      "step": 25800
    },
    {
      "epoch": 1.1681794517757305,
      "grad_norm": 0.5364841222763062,
      "learning_rate": 1.3994468204384504e-05,
      "loss": 0.1289,
      "num_input_tokens_seen": 41510968,
      "step": 25805
    },
    {
      "epoch": 1.1684058036623735,
      "grad_norm": 0.562596321105957,
      "learning_rate": 1.398565404541622e-05,
      "loss": 0.162,
      "num_input_tokens_seen": 41519128,
      "step": 25810
    },
    {
      "epoch": 1.1686321555490164,
      "grad_norm": 0.7453277111053467,
      "learning_rate": 1.3976841584998513e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 41526744,
      "step": 25815
    },
    {
      "epoch": 1.1688585074356594,
      "grad_norm": 0.4336937367916107,
      "learning_rate": 1.3968030824490352e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 41534776,
      "step": 25820
    },
    {
      "epoch": 1.1690848593223024,
      "grad_norm": 0.8249770998954773,
      "learning_rate": 1.3959221765250469e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 41543384,
      "step": 25825
    },
    {
      "epoch": 1.1693112112089454,
      "grad_norm": 0.4713785946369171,
      "learning_rate": 1.3950414408637343e-05,
      "loss": 0.1564,
      "num_input_tokens_seen": 41551672,
      "step": 25830
    },
    {
      "epoch": 1.1695375630955884,
      "grad_norm": 1.0692108869552612,
      "learning_rate": 1.3941608756009166e-05,
      "loss": 0.1266,
      "num_input_tokens_seen": 41559736,
      "step": 25835
    },
    {
      "epoch": 1.1697639149822314,
      "grad_norm": 0.5170444846153259,
      "learning_rate": 1.3932804808723898e-05,
      "loss": 0.1908,
      "num_input_tokens_seen": 41568312,
      "step": 25840
    },
    {
      "epoch": 1.1699902668688744,
      "grad_norm": 1.1199030876159668,
      "learning_rate": 1.3924002568139194e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 41576248,
      "step": 25845
    },
    {
      "epoch": 1.1702166187555174,
      "grad_norm": 1.1819050312042236,
      "learning_rate": 1.3915202035612485e-05,
      "loss": 0.1678,
      "num_input_tokens_seen": 41584056,
      "step": 25850
    },
    {
      "epoch": 1.1704429706421604,
      "grad_norm": 0.30737343430519104,
      "learning_rate": 1.3906403212500935e-05,
      "loss": 0.1203,
      "num_input_tokens_seen": 41591992,
      "step": 25855
    },
    {
      "epoch": 1.1706693225288034,
      "grad_norm": 0.8830257654190063,
      "learning_rate": 1.3897606100161409e-05,
      "loss": 0.1498,
      "num_input_tokens_seen": 41599896,
      "step": 25860
    },
    {
      "epoch": 1.1708956744154462,
      "grad_norm": 0.6792660355567932,
      "learning_rate": 1.388881069995055e-05,
      "loss": 0.1409,
      "num_input_tokens_seen": 41607288,
      "step": 25865
    },
    {
      "epoch": 1.1711220263020892,
      "grad_norm": 0.8364369869232178,
      "learning_rate": 1.3880017013224708e-05,
      "loss": 0.1388,
      "num_input_tokens_seen": 41615224,
      "step": 25870
    },
    {
      "epoch": 1.1713483781887322,
      "grad_norm": 0.9380712509155273,
      "learning_rate": 1.3871225041339984e-05,
      "loss": 0.1873,
      "num_input_tokens_seen": 41623128,
      "step": 25875
    },
    {
      "epoch": 1.1715747300753752,
      "grad_norm": 0.296146035194397,
      "learning_rate": 1.386243478565222e-05,
      "loss": 0.1061,
      "num_input_tokens_seen": 41631448,
      "step": 25880
    },
    {
      "epoch": 1.1718010819620182,
      "grad_norm": 0.6059318780899048,
      "learning_rate": 1.3853646247516966e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 41639064,
      "step": 25885
    },
    {
      "epoch": 1.1720274338486611,
      "grad_norm": 1.0501625537872314,
      "learning_rate": 1.3844859428289545e-05,
      "loss": 0.1542,
      "num_input_tokens_seen": 41647192,
      "step": 25890
    },
    {
      "epoch": 1.1722537857353041,
      "grad_norm": 0.4321511387825012,
      "learning_rate": 1.3836074329324984e-05,
      "loss": 0.1698,
      "num_input_tokens_seen": 41655192,
      "step": 25895
    },
    {
      "epoch": 1.1724801376219471,
      "grad_norm": 1.1021827459335327,
      "learning_rate": 1.3827290951978044e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 41663288,
      "step": 25900
    },
    {
      "epoch": 1.1727064895085901,
      "grad_norm": 0.9208128452301025,
      "learning_rate": 1.381850929760326e-05,
      "loss": 0.1161,
      "num_input_tokens_seen": 41671704,
      "step": 25905
    },
    {
      "epoch": 1.172932841395233,
      "grad_norm": 0.49619120359420776,
      "learning_rate": 1.3809729367554842e-05,
      "loss": 0.1446,
      "num_input_tokens_seen": 41679928,
      "step": 25910
    },
    {
      "epoch": 1.173159193281876,
      "grad_norm": 0.3255995512008667,
      "learning_rate": 1.3800951163186784e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 41687800,
      "step": 25915
    },
    {
      "epoch": 1.173385545168519,
      "grad_norm": 0.801944375038147,
      "learning_rate": 1.3792174685852801e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 41695896,
      "step": 25920
    },
    {
      "epoch": 1.173611897055162,
      "grad_norm": 0.7863603830337524,
      "learning_rate": 1.378339993690632e-05,
      "loss": 0.2055,
      "num_input_tokens_seen": 41703608,
      "step": 25925
    },
    {
      "epoch": 1.1738382489418049,
      "grad_norm": 0.6641141176223755,
      "learning_rate": 1.3774626917700523e-05,
      "loss": 0.1,
      "num_input_tokens_seen": 41711416,
      "step": 25930
    },
    {
      "epoch": 1.1740646008284479,
      "grad_norm": 1.1144696474075317,
      "learning_rate": 1.3765855629588334e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 41719608,
      "step": 25935
    },
    {
      "epoch": 1.1742909527150909,
      "grad_norm": 0.5337851643562317,
      "learning_rate": 1.3757086073922374e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 41727864,
      "step": 25940
    },
    {
      "epoch": 1.1745173046017339,
      "grad_norm": 0.6373708844184875,
      "learning_rate": 1.3748318252055038e-05,
      "loss": 0.1176,
      "num_input_tokens_seen": 41736088,
      "step": 25945
    },
    {
      "epoch": 1.1747436564883769,
      "grad_norm": 0.8261509537696838,
      "learning_rate": 1.3739552165338416e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 41744440,
      "step": 25950
    },
    {
      "epoch": 1.1749700083750199,
      "grad_norm": 0.40077483654022217,
      "learning_rate": 1.3730787815124354e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 41752024,
      "step": 25955
    },
    {
      "epoch": 1.1751963602616629,
      "grad_norm": 0.37123432755470276,
      "learning_rate": 1.3722025202764443e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 41760120,
      "step": 25960
    },
    {
      "epoch": 1.1754227121483058,
      "grad_norm": 0.3691679537296295,
      "learning_rate": 1.371326432960997e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 41767896,
      "step": 25965
    },
    {
      "epoch": 1.1756490640349488,
      "grad_norm": 0.7140098214149475,
      "learning_rate": 1.3704505197011969e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 41775768,
      "step": 25970
    },
    {
      "epoch": 1.1758754159215916,
      "grad_norm": 0.5775566697120667,
      "learning_rate": 1.3695747806321224e-05,
      "loss": 0.1361,
      "num_input_tokens_seen": 41784440,
      "step": 25975
    },
    {
      "epoch": 1.1761017678082346,
      "grad_norm": 0.2489263415336609,
      "learning_rate": 1.3686992158888212e-05,
      "loss": 0.1197,
      "num_input_tokens_seen": 41792216,
      "step": 25980
    },
    {
      "epoch": 1.1763281196948776,
      "grad_norm": 0.4863775968551636,
      "learning_rate": 1.367823825606319e-05,
      "loss": 0.1271,
      "num_input_tokens_seen": 41800344,
      "step": 25985
    },
    {
      "epoch": 1.1765544715815206,
      "grad_norm": 0.4829511344432831,
      "learning_rate": 1.36694860991961e-05,
      "loss": 0.1548,
      "num_input_tokens_seen": 41808344,
      "step": 25990
    },
    {
      "epoch": 1.1767808234681636,
      "grad_norm": 0.8974192142486572,
      "learning_rate": 1.3660735689636636e-05,
      "loss": 0.1343,
      "num_input_tokens_seen": 41816472,
      "step": 25995
    },
    {
      "epoch": 1.1770071753548066,
      "grad_norm": 0.266567587852478,
      "learning_rate": 1.365198702873424e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 41824280,
      "step": 26000
    },
    {
      "epoch": 1.1770071753548066,
      "eval_loss": 0.14456883072853088,
      "eval_runtime": 404.3693,
      "eval_samples_per_second": 97.117,
      "eval_steps_per_second": 24.28,
      "num_input_tokens_seen": 41824280,
      "step": 26000
    },
    {
      "epoch": 1.1772335272414496,
      "grad_norm": 0.6709932088851929,
      "learning_rate": 1.364324011783804e-05,
      "loss": 0.1351,
      "num_input_tokens_seen": 41831992,
      "step": 26005
    },
    {
      "epoch": 1.1774598791280926,
      "grad_norm": 0.7510812878608704,
      "learning_rate": 1.3634494958296934e-05,
      "loss": 0.1378,
      "num_input_tokens_seen": 41839704,
      "step": 26010
    },
    {
      "epoch": 1.1776862310147356,
      "grad_norm": 0.8613909482955933,
      "learning_rate": 1.3625751551459542e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 41847352,
      "step": 26015
    },
    {
      "epoch": 1.1779125829013786,
      "grad_norm": 0.7578063011169434,
      "learning_rate": 1.3617009898674188e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 41856184,
      "step": 26020
    },
    {
      "epoch": 1.1781389347880213,
      "grad_norm": 1.0276005268096924,
      "learning_rate": 1.3608270001288967e-05,
      "loss": 0.15,
      "num_input_tokens_seen": 41863640,
      "step": 26025
    },
    {
      "epoch": 1.1783652866746643,
      "grad_norm": 0.8766303658485413,
      "learning_rate": 1.359953186065166e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 41871192,
      "step": 26030
    },
    {
      "epoch": 1.1785916385613073,
      "grad_norm": 0.7891768217086792,
      "learning_rate": 1.3590795478109814e-05,
      "loss": 0.1677,
      "num_input_tokens_seen": 41878936,
      "step": 26035
    },
    {
      "epoch": 1.1788179904479503,
      "grad_norm": 0.5428935885429382,
      "learning_rate": 1.3582060855010675e-05,
      "loss": 0.1435,
      "num_input_tokens_seen": 41886872,
      "step": 26040
    },
    {
      "epoch": 1.1790443423345933,
      "grad_norm": 0.7590482234954834,
      "learning_rate": 1.3573327992701245e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 41894776,
      "step": 26045
    },
    {
      "epoch": 1.1792706942212363,
      "grad_norm": 0.7009126543998718,
      "learning_rate": 1.356459689252823e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 41903352,
      "step": 26050
    },
    {
      "epoch": 1.1794970461078793,
      "grad_norm": 1.2542450428009033,
      "learning_rate": 1.3555867555838087e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 41910936,
      "step": 26055
    },
    {
      "epoch": 1.1797233979945223,
      "grad_norm": 0.8379874229431152,
      "learning_rate": 1.3547139983976975e-05,
      "loss": 0.1627,
      "num_input_tokens_seen": 41919192,
      "step": 26060
    },
    {
      "epoch": 1.1799497498811653,
      "grad_norm": 0.6278911232948303,
      "learning_rate": 1.3538414178290815e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 41927736,
      "step": 26065
    },
    {
      "epoch": 1.1801761017678083,
      "grad_norm": 0.2813248336315155,
      "learning_rate": 1.3529690140125209e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 41935416,
      "step": 26070
    },
    {
      "epoch": 1.1804024536544513,
      "grad_norm": 1.1760765314102173,
      "learning_rate": 1.352096787082553e-05,
      "loss": 0.1735,
      "num_input_tokens_seen": 41943096,
      "step": 26075
    },
    {
      "epoch": 1.1806288055410943,
      "grad_norm": 0.8599151372909546,
      "learning_rate": 1.3512247371736871e-05,
      "loss": 0.1573,
      "num_input_tokens_seen": 41951864,
      "step": 26080
    },
    {
      "epoch": 1.1808551574277373,
      "grad_norm": 1.2177318334579468,
      "learning_rate": 1.3503528644204022e-05,
      "loss": 0.1421,
      "num_input_tokens_seen": 41960888,
      "step": 26085
    },
    {
      "epoch": 1.18108150931438,
      "grad_norm": 0.884499728679657,
      "learning_rate": 1.349481168957153e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 41968888,
      "step": 26090
    },
    {
      "epoch": 1.181307861201023,
      "grad_norm": 0.6605002284049988,
      "learning_rate": 1.3486096509183665e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 41976984,
      "step": 26095
    },
    {
      "epoch": 1.181534213087666,
      "grad_norm": 0.8300495147705078,
      "learning_rate": 1.3477383104384406e-05,
      "loss": 0.1781,
      "num_input_tokens_seen": 41984824,
      "step": 26100
    },
    {
      "epoch": 1.181760564974309,
      "grad_norm": 0.42988619208335876,
      "learning_rate": 1.3468671476517481e-05,
      "loss": 0.1362,
      "num_input_tokens_seen": 41992888,
      "step": 26105
    },
    {
      "epoch": 1.181986916860952,
      "grad_norm": 0.6954816579818726,
      "learning_rate": 1.3459961626926326e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 42001272,
      "step": 26110
    },
    {
      "epoch": 1.182213268747595,
      "grad_norm": 0.6575515866279602,
      "learning_rate": 1.3451253556954101e-05,
      "loss": 0.1079,
      "num_input_tokens_seen": 42009016,
      "step": 26115
    },
    {
      "epoch": 1.182439620634238,
      "grad_norm": 0.547781765460968,
      "learning_rate": 1.3442547267943717e-05,
      "loss": 0.1319,
      "num_input_tokens_seen": 42016824,
      "step": 26120
    },
    {
      "epoch": 1.182665972520881,
      "grad_norm": 0.37158864736557007,
      "learning_rate": 1.3433842761237774e-05,
      "loss": 0.1211,
      "num_input_tokens_seen": 42025112,
      "step": 26125
    },
    {
      "epoch": 1.182892324407524,
      "grad_norm": 0.45777615904808044,
      "learning_rate": 1.3425140038178639e-05,
      "loss": 0.1195,
      "num_input_tokens_seen": 42033880,
      "step": 26130
    },
    {
      "epoch": 1.183118676294167,
      "grad_norm": 0.7205997109413147,
      "learning_rate": 1.3416439100108358e-05,
      "loss": 0.174,
      "num_input_tokens_seen": 42042104,
      "step": 26135
    },
    {
      "epoch": 1.1833450281808098,
      "grad_norm": 0.3451671302318573,
      "learning_rate": 1.3407739948368734e-05,
      "loss": 0.1531,
      "num_input_tokens_seen": 42050424,
      "step": 26140
    },
    {
      "epoch": 1.1835713800674528,
      "grad_norm": 0.4232098162174225,
      "learning_rate": 1.3399042584301298e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 42058648,
      "step": 26145
    },
    {
      "epoch": 1.1837977319540958,
      "grad_norm": 0.42560893297195435,
      "learning_rate": 1.3390347009247272e-05,
      "loss": 0.1291,
      "num_input_tokens_seen": 42066712,
      "step": 26150
    },
    {
      "epoch": 1.1840240838407388,
      "grad_norm": 0.8784289956092834,
      "learning_rate": 1.3381653224547635e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 42075032,
      "step": 26155
    },
    {
      "epoch": 1.1842504357273818,
      "grad_norm": 0.7234224677085876,
      "learning_rate": 1.3372961231543086e-05,
      "loss": 0.1773,
      "num_input_tokens_seen": 42082808,
      "step": 26160
    },
    {
      "epoch": 1.1844767876140248,
      "grad_norm": 1.0997790098190308,
      "learning_rate": 1.3364271031574016e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 42091800,
      "step": 26165
    },
    {
      "epoch": 1.1847031395006677,
      "grad_norm": 0.39238324761390686,
      "learning_rate": 1.335558262598059e-05,
      "loss": 0.1599,
      "num_input_tokens_seen": 42100120,
      "step": 26170
    },
    {
      "epoch": 1.1849294913873107,
      "grad_norm": 0.7045710682868958,
      "learning_rate": 1.3346896016102645e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 42108248,
      "step": 26175
    },
    {
      "epoch": 1.1851558432739537,
      "grad_norm": 0.9581435322761536,
      "learning_rate": 1.3338211203279788e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 42117304,
      "step": 26180
    },
    {
      "epoch": 1.1853821951605967,
      "grad_norm": 0.5258729457855225,
      "learning_rate": 1.3329528188851303e-05,
      "loss": 0.1905,
      "num_input_tokens_seen": 42124856,
      "step": 26185
    },
    {
      "epoch": 1.1856085470472397,
      "grad_norm": 1.1430985927581787,
      "learning_rate": 1.3320846974156242e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 42132920,
      "step": 26190
    },
    {
      "epoch": 1.1858348989338827,
      "grad_norm": 0.4011263847351074,
      "learning_rate": 1.3312167560533337e-05,
      "loss": 0.1115,
      "num_input_tokens_seen": 42140568,
      "step": 26195
    },
    {
      "epoch": 1.1860612508205255,
      "grad_norm": 0.47714903950691223,
      "learning_rate": 1.3303489949321082e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 42148504,
      "step": 26200
    },
    {
      "epoch": 1.1860612508205255,
      "eval_loss": 0.14476770162582397,
      "eval_runtime": 405.1874,
      "eval_samples_per_second": 96.921,
      "eval_steps_per_second": 24.231,
      "num_input_tokens_seen": 42148504,
      "step": 26200
    },
    {
      "epoch": 1.1862876027071685,
      "grad_norm": 0.47680485248565674,
      "learning_rate": 1.3294814141857653e-05,
      "loss": 0.1497,
      "num_input_tokens_seen": 42156088,
      "step": 26205
    },
    {
      "epoch": 1.1865139545938115,
      "grad_norm": 0.737922191619873,
      "learning_rate": 1.3286140139480992e-05,
      "loss": 0.171,
      "num_input_tokens_seen": 42163896,
      "step": 26210
    },
    {
      "epoch": 1.1867403064804545,
      "grad_norm": 1.1469953060150146,
      "learning_rate": 1.3277467943528719e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 42171768,
      "step": 26215
    },
    {
      "epoch": 1.1869666583670975,
      "grad_norm": 0.7876175045967102,
      "learning_rate": 1.3268797555338203e-05,
      "loss": 0.1902,
      "num_input_tokens_seen": 42180056,
      "step": 26220
    },
    {
      "epoch": 1.1871930102537405,
      "grad_norm": 0.5676473379135132,
      "learning_rate": 1.3260128976246533e-05,
      "loss": 0.1373,
      "num_input_tokens_seen": 42188088,
      "step": 26225
    },
    {
      "epoch": 1.1874193621403835,
      "grad_norm": 0.6869552135467529,
      "learning_rate": 1.32514622075905e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 42196024,
      "step": 26230
    },
    {
      "epoch": 1.1876457140270265,
      "grad_norm": 0.3034117817878723,
      "learning_rate": 1.3242797250706638e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 42203576,
      "step": 26235
    },
    {
      "epoch": 1.1878720659136695,
      "grad_norm": 0.35733139514923096,
      "learning_rate": 1.3234134106931195e-05,
      "loss": 0.1335,
      "num_input_tokens_seen": 42211224,
      "step": 26240
    },
    {
      "epoch": 1.1880984178003124,
      "grad_norm": 0.9847234487533569,
      "learning_rate": 1.322547277760013e-05,
      "loss": 0.1826,
      "num_input_tokens_seen": 42219864,
      "step": 26245
    },
    {
      "epoch": 1.1883247696869552,
      "grad_norm": 0.9354339241981506,
      "learning_rate": 1.3216813264049132e-05,
      "loss": 0.1135,
      "num_input_tokens_seen": 42227928,
      "step": 26250
    },
    {
      "epoch": 1.1885511215735982,
      "grad_norm": 0.506628692150116,
      "learning_rate": 1.32081555676136e-05,
      "loss": 0.1692,
      "num_input_tokens_seen": 42235960,
      "step": 26255
    },
    {
      "epoch": 1.1887774734602412,
      "grad_norm": 0.36767998337745667,
      "learning_rate": 1.3199499689628674e-05,
      "loss": 0.1532,
      "num_input_tokens_seen": 42243640,
      "step": 26260
    },
    {
      "epoch": 1.1890038253468842,
      "grad_norm": 0.4225943684577942,
      "learning_rate": 1.3190845631429192e-05,
      "loss": 0.1358,
      "num_input_tokens_seen": 42252056,
      "step": 26265
    },
    {
      "epoch": 1.1892301772335272,
      "grad_norm": 0.5082213878631592,
      "learning_rate": 1.3182193394349704e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 42260344,
      "step": 26270
    },
    {
      "epoch": 1.1894565291201702,
      "grad_norm": 0.5291304588317871,
      "learning_rate": 1.3173542979724507e-05,
      "loss": 0.1749,
      "num_input_tokens_seen": 42268696,
      "step": 26275
    },
    {
      "epoch": 1.1896828810068132,
      "grad_norm": 0.36736318469047546,
      "learning_rate": 1.3164894388887617e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 42276344,
      "step": 26280
    },
    {
      "epoch": 1.1899092328934562,
      "grad_norm": 0.5004573464393616,
      "learning_rate": 1.3156247623172727e-05,
      "loss": 0.1836,
      "num_input_tokens_seen": 42284248,
      "step": 26285
    },
    {
      "epoch": 1.1901355847800992,
      "grad_norm": 0.7663684487342834,
      "learning_rate": 1.3147602683913302e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 42292408,
      "step": 26290
    },
    {
      "epoch": 1.1903619366667422,
      "grad_norm": 1.148407220840454,
      "learning_rate": 1.3138959572442481e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 42300504,
      "step": 26295
    },
    {
      "epoch": 1.1905882885533852,
      "grad_norm": 0.826692521572113,
      "learning_rate": 1.3130318290093146e-05,
      "loss": 0.1503,
      "num_input_tokens_seen": 42308120,
      "step": 26300
    },
    {
      "epoch": 1.1908146404400282,
      "grad_norm": 0.952592134475708,
      "learning_rate": 1.3121678838197909e-05,
      "loss": 0.1605,
      "num_input_tokens_seen": 42315960,
      "step": 26305
    },
    {
      "epoch": 1.1910409923266712,
      "grad_norm": 1.2024604082107544,
      "learning_rate": 1.3113041218089056e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 42323768,
      "step": 26310
    },
    {
      "epoch": 1.191267344213314,
      "grad_norm": 0.6448525786399841,
      "learning_rate": 1.3104405431098626e-05,
      "loss": 0.1637,
      "num_input_tokens_seen": 42331736,
      "step": 26315
    },
    {
      "epoch": 1.191493696099957,
      "grad_norm": 0.560110330581665,
      "learning_rate": 1.3095771478558377e-05,
      "loss": 0.121,
      "num_input_tokens_seen": 42339832,
      "step": 26320
    },
    {
      "epoch": 1.1917200479866,
      "grad_norm": 1.0816866159439087,
      "learning_rate": 1.3087139361799766e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 42348152,
      "step": 26325
    },
    {
      "epoch": 1.191946399873243,
      "grad_norm": 0.6235789656639099,
      "learning_rate": 1.3078509082153964e-05,
      "loss": 0.1274,
      "num_input_tokens_seen": 42356280,
      "step": 26330
    },
    {
      "epoch": 1.192172751759886,
      "grad_norm": 0.7733340859413147,
      "learning_rate": 1.3069880640951885e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 42364696,
      "step": 26335
    },
    {
      "epoch": 1.192399103646529,
      "grad_norm": 0.9147234559059143,
      "learning_rate": 1.3061254039524123e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 42372760,
      "step": 26340
    },
    {
      "epoch": 1.192625455533172,
      "grad_norm": 0.5788688659667969,
      "learning_rate": 1.3052629279201028e-05,
      "loss": 0.1386,
      "num_input_tokens_seen": 42380952,
      "step": 26345
    },
    {
      "epoch": 1.192851807419815,
      "grad_norm": 0.8855894207954407,
      "learning_rate": 1.3044006361312633e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 42388760,
      "step": 26350
    },
    {
      "epoch": 1.1930781593064579,
      "grad_norm": 0.6243007779121399,
      "learning_rate": 1.30353852871887e-05,
      "loss": 0.1365,
      "num_input_tokens_seen": 42396344,
      "step": 26355
    },
    {
      "epoch": 1.1933045111931009,
      "grad_norm": 0.6794300675392151,
      "learning_rate": 1.302676605815873e-05,
      "loss": 0.1711,
      "num_input_tokens_seen": 42404888,
      "step": 26360
    },
    {
      "epoch": 1.1935308630797437,
      "grad_norm": 0.9795092940330505,
      "learning_rate": 1.3018148675551884e-05,
      "loss": 0.123,
      "num_input_tokens_seen": 42413368,
      "step": 26365
    },
    {
      "epoch": 1.1937572149663866,
      "grad_norm": 0.6683023571968079,
      "learning_rate": 1.3009533140697094e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 42421464,
      "step": 26370
    },
    {
      "epoch": 1.1939835668530296,
      "grad_norm": 0.8492265343666077,
      "learning_rate": 1.3000919454922966e-05,
      "loss": 0.14,
      "num_input_tokens_seen": 42429368,
      "step": 26375
    },
    {
      "epoch": 1.1942099187396726,
      "grad_norm": 0.5553218126296997,
      "learning_rate": 1.299230761955785e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 42437368,
      "step": 26380
    },
    {
      "epoch": 1.1944362706263156,
      "grad_norm": 0.23374411463737488,
      "learning_rate": 1.2983697635929807e-05,
      "loss": 0.1246,
      "num_input_tokens_seen": 42445144,
      "step": 26385
    },
    {
      "epoch": 1.1946626225129586,
      "grad_norm": 0.5069452524185181,
      "learning_rate": 1.2975089505366584e-05,
      "loss": 0.1184,
      "num_input_tokens_seen": 42453400,
      "step": 26390
    },
    {
      "epoch": 1.1948889743996016,
      "grad_norm": 0.5880177021026611,
      "learning_rate": 1.2966483229195683e-05,
      "loss": 0.1969,
      "num_input_tokens_seen": 42461368,
      "step": 26395
    },
    {
      "epoch": 1.1951153262862446,
      "grad_norm": 0.7103668451309204,
      "learning_rate": 1.2957878808744283e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 42470008,
      "step": 26400
    },
    {
      "epoch": 1.1951153262862446,
      "eval_loss": 0.14503298699855804,
      "eval_runtime": 404.4767,
      "eval_samples_per_second": 97.091,
      "eval_steps_per_second": 24.273,
      "num_input_tokens_seen": 42470008,
      "step": 26400
    },
    {
      "epoch": 1.1953416781728876,
      "grad_norm": 0.44385454058647156,
      "learning_rate": 1.294927624533931e-05,
      "loss": 0.1383,
      "num_input_tokens_seen": 42477976,
      "step": 26405
    },
    {
      "epoch": 1.1955680300595306,
      "grad_norm": 0.4002891778945923,
      "learning_rate": 1.2940675540307378e-05,
      "loss": 0.1142,
      "num_input_tokens_seen": 42486936,
      "step": 26410
    },
    {
      "epoch": 1.1957943819461736,
      "grad_norm": 0.38888195157051086,
      "learning_rate": 1.2932076694974814e-05,
      "loss": 0.1522,
      "num_input_tokens_seen": 42495288,
      "step": 26415
    },
    {
      "epoch": 1.1960207338328166,
      "grad_norm": 0.5549401044845581,
      "learning_rate": 1.2923479710667682e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 42503224,
      "step": 26420
    },
    {
      "epoch": 1.1962470857194596,
      "grad_norm": 0.4467627704143524,
      "learning_rate": 1.2914884588711751e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 42511448,
      "step": 26425
    },
    {
      "epoch": 1.1964734376061024,
      "grad_norm": 0.5216729044914246,
      "learning_rate": 1.2906291330432475e-05,
      "loss": 0.1286,
      "num_input_tokens_seen": 42519224,
      "step": 26430
    },
    {
      "epoch": 1.1966997894927454,
      "grad_norm": 0.6673722267150879,
      "learning_rate": 1.2897699937155055e-05,
      "loss": 0.1387,
      "num_input_tokens_seen": 42527800,
      "step": 26435
    },
    {
      "epoch": 1.1969261413793884,
      "grad_norm": 0.41061797738075256,
      "learning_rate": 1.2889110410204403e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 42535672,
      "step": 26440
    },
    {
      "epoch": 1.1971524932660313,
      "grad_norm": 0.37137457728385925,
      "learning_rate": 1.2880522750905111e-05,
      "loss": 0.117,
      "num_input_tokens_seen": 42543608,
      "step": 26445
    },
    {
      "epoch": 1.1973788451526743,
      "grad_norm": 0.8049605488777161,
      "learning_rate": 1.2871936960581523e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 42552056,
      "step": 26450
    },
    {
      "epoch": 1.1976051970393173,
      "grad_norm": 0.4607718884944916,
      "learning_rate": 1.2863353040557658e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 42560248,
      "step": 26455
    },
    {
      "epoch": 1.1978315489259603,
      "grad_norm": 0.8691207766532898,
      "learning_rate": 1.2854770992157273e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 42567832,
      "step": 26460
    },
    {
      "epoch": 1.1980579008126033,
      "grad_norm": 0.5525789260864258,
      "learning_rate": 1.2846190816703835e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 42575480,
      "step": 26465
    },
    {
      "epoch": 1.1982842526992463,
      "grad_norm": 1.0862162113189697,
      "learning_rate": 1.2837612515520498e-05,
      "loss": 0.1461,
      "num_input_tokens_seen": 42583480,
      "step": 26470
    },
    {
      "epoch": 1.1985106045858893,
      "grad_norm": 0.41198140382766724,
      "learning_rate": 1.2829036089930163e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 42591256,
      "step": 26475
    },
    {
      "epoch": 1.198736956472532,
      "grad_norm": 0.9420742392539978,
      "learning_rate": 1.2820461541255412e-05,
      "loss": 0.1429,
      "num_input_tokens_seen": 42599128,
      "step": 26480
    },
    {
      "epoch": 1.198963308359175,
      "grad_norm": 0.6206908822059631,
      "learning_rate": 1.2811888870818543e-05,
      "loss": 0.1379,
      "num_input_tokens_seen": 42606840,
      "step": 26485
    },
    {
      "epoch": 1.199189660245818,
      "grad_norm": 0.43749669194221497,
      "learning_rate": 1.2803318079941581e-05,
      "loss": 0.1556,
      "num_input_tokens_seen": 42614840,
      "step": 26490
    },
    {
      "epoch": 1.199416012132461,
      "grad_norm": 1.1778355836868286,
      "learning_rate": 1.2794749169946235e-05,
      "loss": 0.1546,
      "num_input_tokens_seen": 42622552,
      "step": 26495
    },
    {
      "epoch": 1.199642364019104,
      "grad_norm": 1.5866632461547852,
      "learning_rate": 1.2786182142153952e-05,
      "loss": 0.1499,
      "num_input_tokens_seen": 42630136,
      "step": 26500
    },
    {
      "epoch": 1.199868715905747,
      "grad_norm": 0.5575076937675476,
      "learning_rate": 1.2777616997885878e-05,
      "loss": 0.1151,
      "num_input_tokens_seen": 42637912,
      "step": 26505
    },
    {
      "epoch": 1.20009506779239,
      "grad_norm": 0.8329854607582092,
      "learning_rate": 1.2769053738462847e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 42645720,
      "step": 26510
    },
    {
      "epoch": 1.200321419679033,
      "grad_norm": 0.8476817011833191,
      "learning_rate": 1.2760492365205434e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 42653912,
      "step": 26515
    },
    {
      "epoch": 1.200547771565676,
      "grad_norm": 0.4678495228290558,
      "learning_rate": 1.2751932879433919e-05,
      "loss": 0.1938,
      "num_input_tokens_seen": 42661624,
      "step": 26520
    },
    {
      "epoch": 1.200774123452319,
      "grad_norm": 0.551138162612915,
      "learning_rate": 1.2743375282468267e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 42669400,
      "step": 26525
    },
    {
      "epoch": 1.201000475338962,
      "grad_norm": 0.6768054962158203,
      "learning_rate": 1.2734819575628182e-05,
      "loss": 0.1657,
      "num_input_tokens_seen": 42676920,
      "step": 26530
    },
    {
      "epoch": 1.201226827225605,
      "grad_norm": 0.6973799467086792,
      "learning_rate": 1.2726265760233039e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 42685112,
      "step": 26535
    },
    {
      "epoch": 1.2014531791122478,
      "grad_norm": 0.8431814312934875,
      "learning_rate": 1.271771383760197e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 42693208,
      "step": 26540
    },
    {
      "epoch": 1.2016795309988908,
      "grad_norm": 0.3472402095794678,
      "learning_rate": 1.2709163809053764e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 42701336,
      "step": 26545
    },
    {
      "epoch": 1.2019058828855338,
      "grad_norm": 0.37603047490119934,
      "learning_rate": 1.2700615675906963e-05,
      "loss": 0.1419,
      "num_input_tokens_seen": 42709592,
      "step": 26550
    },
    {
      "epoch": 1.2021322347721768,
      "grad_norm": 0.7642000317573547,
      "learning_rate": 1.269206943947978e-05,
      "loss": 0.1483,
      "num_input_tokens_seen": 42717528,
      "step": 26555
    },
    {
      "epoch": 1.2023585866588198,
      "grad_norm": 0.7542961835861206,
      "learning_rate": 1.2683525101090177e-05,
      "loss": 0.1451,
      "num_input_tokens_seen": 42725976,
      "step": 26560
    },
    {
      "epoch": 1.2025849385454628,
      "grad_norm": 0.5801111459732056,
      "learning_rate": 1.2674982662055765e-05,
      "loss": 0.1015,
      "num_input_tokens_seen": 42733464,
      "step": 26565
    },
    {
      "epoch": 1.2028112904321058,
      "grad_norm": 0.3797661066055298,
      "learning_rate": 1.2666442123693922e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 42742040,
      "step": 26570
    },
    {
      "epoch": 1.2030376423187488,
      "grad_norm": 0.9107735753059387,
      "learning_rate": 1.265790348732169e-05,
      "loss": 0.1389,
      "num_input_tokens_seen": 42749880,
      "step": 26575
    },
    {
      "epoch": 1.2032639942053918,
      "grad_norm": 0.921610951423645,
      "learning_rate": 1.264936675425584e-05,
      "loss": 0.1528,
      "num_input_tokens_seen": 42757656,
      "step": 26580
    },
    {
      "epoch": 1.2034903460920348,
      "grad_norm": 0.8030891418457031,
      "learning_rate": 1.2640831925812852e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 42765368,
      "step": 26585
    },
    {
      "epoch": 1.2037166979786775,
      "grad_norm": 1.1026256084442139,
      "learning_rate": 1.263229900330889e-05,
      "loss": 0.1656,
      "num_input_tokens_seen": 42773304,
      "step": 26590
    },
    {
      "epoch": 1.2039430498653205,
      "grad_norm": 0.6891993284225464,
      "learning_rate": 1.2623767988059843e-05,
      "loss": 0.1234,
      "num_input_tokens_seen": 42781048,
      "step": 26595
    },
    {
      "epoch": 1.2041694017519635,
      "grad_norm": 1.1785855293273926,
      "learning_rate": 1.2615238881381309e-05,
      "loss": 0.1303,
      "num_input_tokens_seen": 42789400,
      "step": 26600
    },
    {
      "epoch": 1.2041694017519635,
      "eval_loss": 0.1451791375875473,
      "eval_runtime": 404.0317,
      "eval_samples_per_second": 97.198,
      "eval_steps_per_second": 24.3,
      "num_input_tokens_seen": 42789400,
      "step": 26600
    },
    {
      "epoch": 1.2043957536386065,
      "grad_norm": 0.4941191077232361,
      "learning_rate": 1.2606711684588568e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 42797272,
      "step": 26605
    },
    {
      "epoch": 1.2046221055252495,
      "grad_norm": 0.30654484033584595,
      "learning_rate": 1.2598186398996636e-05,
      "loss": 0.1299,
      "num_input_tokens_seen": 42806424,
      "step": 26610
    },
    {
      "epoch": 1.2048484574118925,
      "grad_norm": 0.8263979554176331,
      "learning_rate": 1.2589663025920207e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 42814104,
      "step": 26615
    },
    {
      "epoch": 1.2050748092985355,
      "grad_norm": 0.3511446416378021,
      "learning_rate": 1.2581141566673705e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 42822392,
      "step": 26620
    },
    {
      "epoch": 1.2053011611851785,
      "grad_norm": 0.4533085525035858,
      "learning_rate": 1.257262202257124e-05,
      "loss": 0.139,
      "num_input_tokens_seen": 42830040,
      "step": 26625
    },
    {
      "epoch": 1.2055275130718215,
      "grad_norm": 0.4968113303184509,
      "learning_rate": 1.2564104394926618e-05,
      "loss": 0.1613,
      "num_input_tokens_seen": 42837784,
      "step": 26630
    },
    {
      "epoch": 1.2057538649584645,
      "grad_norm": 0.5306010842323303,
      "learning_rate": 1.2555588685053383e-05,
      "loss": 0.1283,
      "num_input_tokens_seen": 42845912,
      "step": 26635
    },
    {
      "epoch": 1.2059802168451075,
      "grad_norm": 0.43359509110450745,
      "learning_rate": 1.2547074894264762e-05,
      "loss": 0.1431,
      "num_input_tokens_seen": 42853368,
      "step": 26640
    },
    {
      "epoch": 1.2062065687317505,
      "grad_norm": 0.5020190477371216,
      "learning_rate": 1.2538563023873679e-05,
      "loss": 0.1244,
      "num_input_tokens_seen": 42861944,
      "step": 26645
    },
    {
      "epoch": 1.2064329206183935,
      "grad_norm": 0.6085378527641296,
      "learning_rate": 1.2530053075192789e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 42870072,
      "step": 26650
    },
    {
      "epoch": 1.2066592725050362,
      "grad_norm": 0.5097737312316895,
      "learning_rate": 1.252154504953441e-05,
      "loss": 0.1032,
      "num_input_tokens_seen": 42877816,
      "step": 26655
    },
    {
      "epoch": 1.2068856243916792,
      "grad_norm": 0.8386852145195007,
      "learning_rate": 1.25130389482106e-05,
      "loss": 0.1368,
      "num_input_tokens_seen": 42885720,
      "step": 26660
    },
    {
      "epoch": 1.2071119762783222,
      "grad_norm": 0.6063821911811829,
      "learning_rate": 1.2504534772533116e-05,
      "loss": 0.1233,
      "num_input_tokens_seen": 42893752,
      "step": 26665
    },
    {
      "epoch": 1.2073383281649652,
      "grad_norm": 0.9983562231063843,
      "learning_rate": 1.2496032523813387e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 42901368,
      "step": 26670
    },
    {
      "epoch": 1.2075646800516082,
      "grad_norm": 0.9516387581825256,
      "learning_rate": 1.2487532203362576e-05,
      "loss": 0.1524,
      "num_input_tokens_seen": 42909048,
      "step": 26675
    },
    {
      "epoch": 1.2077910319382512,
      "grad_norm": 1.3894007205963135,
      "learning_rate": 1.247903381249155e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 42917336,
      "step": 26680
    },
    {
      "epoch": 1.2080173838248942,
      "grad_norm": 0.6215218901634216,
      "learning_rate": 1.2470537352510853e-05,
      "loss": 0.0881,
      "num_input_tokens_seen": 42926008,
      "step": 26685
    },
    {
      "epoch": 1.2082437357115372,
      "grad_norm": 0.5240797400474548,
      "learning_rate": 1.2462042824730758e-05,
      "loss": 0.1455,
      "num_input_tokens_seen": 42933976,
      "step": 26690
    },
    {
      "epoch": 1.2084700875981802,
      "grad_norm": 0.4059458374977112,
      "learning_rate": 1.245355023046122e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 42942776,
      "step": 26695
    },
    {
      "epoch": 1.2086964394848232,
      "grad_norm": 1.174546718597412,
      "learning_rate": 1.2445059571011896e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 42950744,
      "step": 26700
    },
    {
      "epoch": 1.208922791371466,
      "grad_norm": 0.7399083375930786,
      "learning_rate": 1.2436570847692173e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 42958328,
      "step": 26705
    },
    {
      "epoch": 1.209149143258109,
      "grad_norm": 0.4135076403617859,
      "learning_rate": 1.2428084061811096e-05,
      "loss": 0.1996,
      "num_input_tokens_seen": 42966328,
      "step": 26710
    },
    {
      "epoch": 1.209375495144752,
      "grad_norm": 0.8373847603797913,
      "learning_rate": 1.2419599214677447e-05,
      "loss": 0.1337,
      "num_input_tokens_seen": 42974392,
      "step": 26715
    },
    {
      "epoch": 1.209601847031395,
      "grad_norm": 0.5701278448104858,
      "learning_rate": 1.2411116307599702e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 42982424,
      "step": 26720
    },
    {
      "epoch": 1.209828198918038,
      "grad_norm": 1.3670107126235962,
      "learning_rate": 1.2402635341886016e-05,
      "loss": 0.1774,
      "num_input_tokens_seen": 42990360,
      "step": 26725
    },
    {
      "epoch": 1.210054550804681,
      "grad_norm": 1.5741297006607056,
      "learning_rate": 1.2394156318844278e-05,
      "loss": 0.144,
      "num_input_tokens_seen": 42998616,
      "step": 26730
    },
    {
      "epoch": 1.210280902691324,
      "grad_norm": 0.4259689748287201,
      "learning_rate": 1.2385679239782039e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 43006008,
      "step": 26735
    },
    {
      "epoch": 1.210507254577967,
      "grad_norm": 1.12558913230896,
      "learning_rate": 1.2377204106006585e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 43013752,
      "step": 26740
    },
    {
      "epoch": 1.21073360646461,
      "grad_norm": 0.6209205389022827,
      "learning_rate": 1.2368730918824891e-05,
      "loss": 0.1618,
      "num_input_tokens_seen": 43021912,
      "step": 26745
    },
    {
      "epoch": 1.210959958351253,
      "grad_norm": 0.874366819858551,
      "learning_rate": 1.236025967954362e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 43029752,
      "step": 26750
    },
    {
      "epoch": 1.211186310237896,
      "grad_norm": 0.5971819162368774,
      "learning_rate": 1.2351790389469153e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 43037624,
      "step": 26755
    },
    {
      "epoch": 1.211412662124539,
      "grad_norm": 0.7555791735649109,
      "learning_rate": 1.234332304990755e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 43045592,
      "step": 26760
    },
    {
      "epoch": 1.211639014011182,
      "grad_norm": 0.9936678409576416,
      "learning_rate": 1.2334857662164593e-05,
      "loss": 0.1661,
      "num_input_tokens_seen": 43053432,
      "step": 26765
    },
    {
      "epoch": 1.2118653658978247,
      "grad_norm": 0.6895809173583984,
      "learning_rate": 1.2326394227545743e-05,
      "loss": 0.1491,
      "num_input_tokens_seen": 43061464,
      "step": 26770
    },
    {
      "epoch": 1.2120917177844677,
      "grad_norm": 0.7341305017471313,
      "learning_rate": 1.2317932747356162e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 43069112,
      "step": 26775
    },
    {
      "epoch": 1.2123180696711107,
      "grad_norm": 0.6403290629386902,
      "learning_rate": 1.2309473222900726e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 43077176,
      "step": 26780
    },
    {
      "epoch": 1.2125444215577537,
      "grad_norm": 0.5113081336021423,
      "learning_rate": 1.2301015655484006e-05,
      "loss": 0.1703,
      "num_input_tokens_seen": 43084792,
      "step": 26785
    },
    {
      "epoch": 1.2127707734443967,
      "grad_norm": 0.7808510065078735,
      "learning_rate": 1.2292560046410245e-05,
      "loss": 0.1265,
      "num_input_tokens_seen": 43093048,
      "step": 26790
    },
    {
      "epoch": 1.2129971253310396,
      "grad_norm": 0.6356838941574097,
      "learning_rate": 1.228410639698343e-05,
      "loss": 0.191,
      "num_input_tokens_seen": 43101272,
      "step": 26795
    },
    {
      "epoch": 1.2132234772176826,
      "grad_norm": 1.3385300636291504,
      "learning_rate": 1.2275654708507195e-05,
      "loss": 0.1889,
      "num_input_tokens_seen": 43109688,
      "step": 26800
    },
    {
      "epoch": 1.2132234772176826,
      "eval_loss": 0.1449344903230667,
      "eval_runtime": 404.668,
      "eval_samples_per_second": 97.045,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 43109688,
      "step": 26800
    },
    {
      "epoch": 1.2134498291043256,
      "grad_norm": 1.0991743803024292,
      "learning_rate": 1.2267204982284908e-05,
      "loss": 0.191,
      "num_input_tokens_seen": 43118776,
      "step": 26805
    },
    {
      "epoch": 1.2136761809909686,
      "grad_norm": 0.34008243680000305,
      "learning_rate": 1.2258757219619635e-05,
      "loss": 0.1364,
      "num_input_tokens_seen": 43127192,
      "step": 26810
    },
    {
      "epoch": 1.2139025328776116,
      "grad_norm": 0.6945281028747559,
      "learning_rate": 1.2250311421814104e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 43135064,
      "step": 26815
    },
    {
      "epoch": 1.2141288847642544,
      "grad_norm": 0.9258999228477478,
      "learning_rate": 1.2241867590170772e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 43143288,
      "step": 26820
    },
    {
      "epoch": 1.2143552366508974,
      "grad_norm": 1.3187055587768555,
      "learning_rate": 1.2233425725991799e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 43150744,
      "step": 26825
    },
    {
      "epoch": 1.2145815885375404,
      "grad_norm": 0.4694218039512634,
      "learning_rate": 1.2224985830579003e-05,
      "loss": 0.125,
      "num_input_tokens_seen": 43160344,
      "step": 26830
    },
    {
      "epoch": 1.2148079404241834,
      "grad_norm": 0.5555428862571716,
      "learning_rate": 1.2216547905233944e-05,
      "loss": 0.1118,
      "num_input_tokens_seen": 43167864,
      "step": 26835
    },
    {
      "epoch": 1.2150342923108264,
      "grad_norm": 0.8058848977088928,
      "learning_rate": 1.2208111951257842e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 43176696,
      "step": 26840
    },
    {
      "epoch": 1.2152606441974694,
      "grad_norm": 0.5512514710426331,
      "learning_rate": 1.2199677969951622e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 43184504,
      "step": 26845
    },
    {
      "epoch": 1.2154869960841124,
      "grad_norm": 0.2977699339389801,
      "learning_rate": 1.2191245962615927e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 43192632,
      "step": 26850
    },
    {
      "epoch": 1.2157133479707554,
      "grad_norm": 0.9622696042060852,
      "learning_rate": 1.218281593055106e-05,
      "loss": 0.1353,
      "num_input_tokens_seen": 43201112,
      "step": 26855
    },
    {
      "epoch": 1.2159396998573984,
      "grad_norm": 0.7683820128440857,
      "learning_rate": 1.217438787505705e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 43208952,
      "step": 26860
    },
    {
      "epoch": 1.2161660517440414,
      "grad_norm": 0.5446147322654724,
      "learning_rate": 1.2165961797433615e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 43217016,
      "step": 26865
    },
    {
      "epoch": 1.2163924036306843,
      "grad_norm": 0.7341500520706177,
      "learning_rate": 1.215753769898014e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 43224568,
      "step": 26870
    },
    {
      "epoch": 1.2166187555173273,
      "grad_norm": 0.6026288270950317,
      "learning_rate": 1.2149115580995755e-05,
      "loss": 0.101,
      "num_input_tokens_seen": 43232600,
      "step": 26875
    },
    {
      "epoch": 1.2168451074039701,
      "grad_norm": 0.367372065782547,
      "learning_rate": 1.2140695444779227e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 43240568,
      "step": 26880
    },
    {
      "epoch": 1.2170714592906131,
      "grad_norm": 0.7635295987129211,
      "learning_rate": 1.2132277291629066e-05,
      "loss": 0.1571,
      "num_input_tokens_seen": 43248664,
      "step": 26885
    },
    {
      "epoch": 1.217297811177256,
      "grad_norm": 1.5843735933303833,
      "learning_rate": 1.2123861122843458e-05,
      "loss": 0.134,
      "num_input_tokens_seen": 43256888,
      "step": 26890
    },
    {
      "epoch": 1.217524163063899,
      "grad_norm": 0.4856220781803131,
      "learning_rate": 1.2115446939720271e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 43264792,
      "step": 26895
    },
    {
      "epoch": 1.217750514950542,
      "grad_norm": 0.7626066207885742,
      "learning_rate": 1.210703474355708e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 43273176,
      "step": 26900
    },
    {
      "epoch": 1.217976866837185,
      "grad_norm": 0.7281409502029419,
      "learning_rate": 1.2098624535651164e-05,
      "loss": 0.1504,
      "num_input_tokens_seen": 43280888,
      "step": 26905
    },
    {
      "epoch": 1.218203218723828,
      "grad_norm": 1.1819367408752441,
      "learning_rate": 1.2090216317299477e-05,
      "loss": 0.1054,
      "num_input_tokens_seen": 43288600,
      "step": 26910
    },
    {
      "epoch": 1.218429570610471,
      "grad_norm": 0.8371464014053345,
      "learning_rate": 1.2081810089798668e-05,
      "loss": 0.1849,
      "num_input_tokens_seen": 43296088,
      "step": 26915
    },
    {
      "epoch": 1.218655922497114,
      "grad_norm": 0.7800661325454712,
      "learning_rate": 1.2073405854445072e-05,
      "loss": 0.1695,
      "num_input_tokens_seen": 43303768,
      "step": 26920
    },
    {
      "epoch": 1.218882274383757,
      "grad_norm": 0.5044596195220947,
      "learning_rate": 1.206500361253474e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 43311608,
      "step": 26925
    },
    {
      "epoch": 1.2191086262703998,
      "grad_norm": 0.6589491367340088,
      "learning_rate": 1.2056603365363409e-05,
      "loss": 0.1433,
      "num_input_tokens_seen": 43320280,
      "step": 26930
    },
    {
      "epoch": 1.2193349781570428,
      "grad_norm": 0.6000413298606873,
      "learning_rate": 1.2048205114226487e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 43328824,
      "step": 26935
    },
    {
      "epoch": 1.2195613300436858,
      "grad_norm": 0.9536099433898926,
      "learning_rate": 1.2039808860419102e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 43336504,
      "step": 26940
    },
    {
      "epoch": 1.2197876819303288,
      "grad_norm": 0.4389052093029022,
      "learning_rate": 1.2031414605236066e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 43344824,
      "step": 26945
    },
    {
      "epoch": 1.2200140338169718,
      "grad_norm": 0.694267988204956,
      "learning_rate": 1.2023022349971862e-05,
      "loss": 0.1474,
      "num_input_tokens_seen": 43353240,
      "step": 26950
    },
    {
      "epoch": 1.2202403857036148,
      "grad_norm": 0.5810641050338745,
      "learning_rate": 1.20146320959207e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 43361912,
      "step": 26955
    },
    {
      "epoch": 1.2204667375902578,
      "grad_norm": 0.7351385354995728,
      "learning_rate": 1.2006243844376445e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 43369912,
      "step": 26960
    },
    {
      "epoch": 1.2206930894769008,
      "grad_norm": 0.6888794302940369,
      "learning_rate": 1.1997857596632678e-05,
      "loss": 0.1672,
      "num_input_tokens_seen": 43377944,
      "step": 26965
    },
    {
      "epoch": 1.2209194413635438,
      "grad_norm": 0.7496941089630127,
      "learning_rate": 1.1989473353982672e-05,
      "loss": 0.1572,
      "num_input_tokens_seen": 43386424,
      "step": 26970
    },
    {
      "epoch": 1.2211457932501868,
      "grad_norm": 0.8782442212104797,
      "learning_rate": 1.198109111771937e-05,
      "loss": 0.0987,
      "num_input_tokens_seen": 43394104,
      "step": 26975
    },
    {
      "epoch": 1.2213721451368298,
      "grad_norm": 0.6522481441497803,
      "learning_rate": 1.197271088913543e-05,
      "loss": 0.1521,
      "num_input_tokens_seen": 43401656,
      "step": 26980
    },
    {
      "epoch": 1.2215984970234728,
      "grad_norm": 0.44367891550064087,
      "learning_rate": 1.1964332669523182e-05,
      "loss": 0.1169,
      "num_input_tokens_seen": 43408952,
      "step": 26985
    },
    {
      "epoch": 1.2218248489101158,
      "grad_norm": 0.9575531482696533,
      "learning_rate": 1.1955956460174645e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 43416664,
      "step": 26990
    },
    {
      "epoch": 1.2220512007967586,
      "grad_norm": 0.6713128685951233,
      "learning_rate": 1.1947582262381552e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 43425144,
      "step": 26995
    },
    {
      "epoch": 1.2222775526834015,
      "grad_norm": 0.8517068028450012,
      "learning_rate": 1.1939210077435293e-05,
      "loss": 0.0996,
      "num_input_tokens_seen": 43433720,
      "step": 27000
    },
    {
      "epoch": 1.2222775526834015,
      "eval_loss": 0.1445118486881256,
      "eval_runtime": 404.2364,
      "eval_samples_per_second": 97.149,
      "eval_steps_per_second": 24.288,
      "num_input_tokens_seen": 43433720,
      "step": 27000
    },
    {
      "epoch": 1.2225039045700445,
      "grad_norm": 0.4993651509284973,
      "learning_rate": 1.193083990662697e-05,
      "loss": 0.112,
      "num_input_tokens_seen": 43442104,
      "step": 27005
    },
    {
      "epoch": 1.2227302564566875,
      "grad_norm": 0.42103490233421326,
      "learning_rate": 1.192247175124738e-05,
      "loss": 0.1111,
      "num_input_tokens_seen": 43449592,
      "step": 27010
    },
    {
      "epoch": 1.2229566083433305,
      "grad_norm": 0.3969486653804779,
      "learning_rate": 1.191410561258698e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 43457144,
      "step": 27015
    },
    {
      "epoch": 1.2231829602299735,
      "grad_norm": 0.4214591383934021,
      "learning_rate": 1.1905741491935944e-05,
      "loss": 0.1254,
      "num_input_tokens_seen": 43465304,
      "step": 27020
    },
    {
      "epoch": 1.2234093121166165,
      "grad_norm": 0.6414113640785217,
      "learning_rate": 1.1897379390584129e-05,
      "loss": 0.1222,
      "num_input_tokens_seen": 43473144,
      "step": 27025
    },
    {
      "epoch": 1.2236356640032595,
      "grad_norm": 0.9869582653045654,
      "learning_rate": 1.1889019309821062e-05,
      "loss": 0.1995,
      "num_input_tokens_seen": 43481848,
      "step": 27030
    },
    {
      "epoch": 1.2238620158899025,
      "grad_norm": 0.35832658410072327,
      "learning_rate": 1.188066125093599e-05,
      "loss": 0.1554,
      "num_input_tokens_seen": 43490488,
      "step": 27035
    },
    {
      "epoch": 1.2240883677765455,
      "grad_norm": 0.4569278955459595,
      "learning_rate": 1.1872305215217811e-05,
      "loss": 0.1371,
      "num_input_tokens_seen": 43498328,
      "step": 27040
    },
    {
      "epoch": 1.2243147196631883,
      "grad_norm": 1.1547908782958984,
      "learning_rate": 1.186395120395514e-05,
      "loss": 0.1696,
      "num_input_tokens_seen": 43506136,
      "step": 27045
    },
    {
      "epoch": 1.2245410715498313,
      "grad_norm": 0.6173463463783264,
      "learning_rate": 1.1855599218436283e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 43514872,
      "step": 27050
    },
    {
      "epoch": 1.2247674234364743,
      "grad_norm": 0.3095342218875885,
      "learning_rate": 1.1847249259949209e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 43522712,
      "step": 27055
    },
    {
      "epoch": 1.2249937753231173,
      "grad_norm": 0.7343573570251465,
      "learning_rate": 1.1838901329781574e-05,
      "loss": 0.1427,
      "num_input_tokens_seen": 43530808,
      "step": 27060
    },
    {
      "epoch": 1.2252201272097603,
      "grad_norm": 0.6459631323814392,
      "learning_rate": 1.1830555429220758e-05,
      "loss": 0.1644,
      "num_input_tokens_seen": 43538552,
      "step": 27065
    },
    {
      "epoch": 1.2254464790964033,
      "grad_norm": 0.37956666946411133,
      "learning_rate": 1.1822211559553784e-05,
      "loss": 0.1218,
      "num_input_tokens_seen": 43546520,
      "step": 27070
    },
    {
      "epoch": 1.2256728309830462,
      "grad_norm": 0.7007660865783691,
      "learning_rate": 1.18138697220674e-05,
      "loss": 0.1139,
      "num_input_tokens_seen": 43554424,
      "step": 27075
    },
    {
      "epoch": 1.2258991828696892,
      "grad_norm": 0.5666101574897766,
      "learning_rate": 1.1805529918048e-05,
      "loss": 0.1355,
      "num_input_tokens_seen": 43562648,
      "step": 27080
    },
    {
      "epoch": 1.2261255347563322,
      "grad_norm": 0.4246196150779724,
      "learning_rate": 1.1797192148781702e-05,
      "loss": 0.0983,
      "num_input_tokens_seen": 43570424,
      "step": 27085
    },
    {
      "epoch": 1.2263518866429752,
      "grad_norm": 0.8911687731742859,
      "learning_rate": 1.1788856415554297e-05,
      "loss": 0.153,
      "num_input_tokens_seen": 43579032,
      "step": 27090
    },
    {
      "epoch": 1.2265782385296182,
      "grad_norm": 0.7579286694526672,
      "learning_rate": 1.1780522719651249e-05,
      "loss": 0.2035,
      "num_input_tokens_seen": 43587288,
      "step": 27095
    },
    {
      "epoch": 1.2268045904162612,
      "grad_norm": 0.7536431550979614,
      "learning_rate": 1.1772191062357721e-05,
      "loss": 0.1418,
      "num_input_tokens_seen": 43595544,
      "step": 27100
    },
    {
      "epoch": 1.2270309423029042,
      "grad_norm": 0.5251725912094116,
      "learning_rate": 1.1763861444958573e-05,
      "loss": 0.1702,
      "num_input_tokens_seen": 43603800,
      "step": 27105
    },
    {
      "epoch": 1.227257294189547,
      "grad_norm": 0.8586909770965576,
      "learning_rate": 1.1755533868738317e-05,
      "loss": 0.1411,
      "num_input_tokens_seen": 43611224,
      "step": 27110
    },
    {
      "epoch": 1.22748364607619,
      "grad_norm": 0.6178776025772095,
      "learning_rate": 1.1747208334981185e-05,
      "loss": 0.1518,
      "num_input_tokens_seen": 43619672,
      "step": 27115
    },
    {
      "epoch": 1.227709997962833,
      "grad_norm": 0.4969874620437622,
      "learning_rate": 1.1738884844971067e-05,
      "loss": 0.1369,
      "num_input_tokens_seen": 43628152,
      "step": 27120
    },
    {
      "epoch": 1.227936349849476,
      "grad_norm": 0.4002637267112732,
      "learning_rate": 1.1730563399991563e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 43635864,
      "step": 27125
    },
    {
      "epoch": 1.228162701736119,
      "grad_norm": 1.3887335062026978,
      "learning_rate": 1.1722244001325938e-05,
      "loss": 0.1356,
      "num_input_tokens_seen": 43643416,
      "step": 27130
    },
    {
      "epoch": 1.228389053622762,
      "grad_norm": 0.7622005939483643,
      "learning_rate": 1.1713926650257137e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 43651160,
      "step": 27135
    },
    {
      "epoch": 1.228615405509405,
      "grad_norm": 1.3950190544128418,
      "learning_rate": 1.170561134806781e-05,
      "loss": 0.1579,
      "num_input_tokens_seen": 43659800,
      "step": 27140
    },
    {
      "epoch": 1.228841757396048,
      "grad_norm": 0.3629354238510132,
      "learning_rate": 1.1697298096040287e-05,
      "loss": 0.1417,
      "num_input_tokens_seen": 43668312,
      "step": 27145
    },
    {
      "epoch": 1.229068109282691,
      "grad_norm": 0.5803984999656677,
      "learning_rate": 1.1688986895456567e-05,
      "loss": 0.1245,
      "num_input_tokens_seen": 43676280,
      "step": 27150
    },
    {
      "epoch": 1.229294461169334,
      "grad_norm": 0.9150949716567993,
      "learning_rate": 1.1680677747598349e-05,
      "loss": 0.1595,
      "num_input_tokens_seen": 43684312,
      "step": 27155
    },
    {
      "epoch": 1.2295208130559767,
      "grad_norm": 1.091056227684021,
      "learning_rate": 1.1672370653746995e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 43692440,
      "step": 27160
    },
    {
      "epoch": 1.2297471649426197,
      "grad_norm": 0.7726839780807495,
      "learning_rate": 1.166406561518357e-05,
      "loss": 0.1399,
      "num_input_tokens_seen": 43700280,
      "step": 27165
    },
    {
      "epoch": 1.2299735168292627,
      "grad_norm": 0.8030665516853333,
      "learning_rate": 1.1655762633188826e-05,
      "loss": 0.1464,
      "num_input_tokens_seen": 43708408,
      "step": 27170
    },
    {
      "epoch": 1.2301998687159057,
      "grad_norm": 0.5702908039093018,
      "learning_rate": 1.1647461709043172e-05,
      "loss": 0.1363,
      "num_input_tokens_seen": 43716248,
      "step": 27175
    },
    {
      "epoch": 1.2304262206025487,
      "grad_norm": 0.9033637642860413,
      "learning_rate": 1.1639162844026722e-05,
      "loss": 0.13,
      "num_input_tokens_seen": 43724088,
      "step": 27180
    },
    {
      "epoch": 1.2306525724891917,
      "grad_norm": 0.8926343321800232,
      "learning_rate": 1.163086603941927e-05,
      "loss": 0.1485,
      "num_input_tokens_seen": 43732024,
      "step": 27185
    },
    {
      "epoch": 1.2308789243758347,
      "grad_norm": 0.6554286479949951,
      "learning_rate": 1.1622571296500273e-05,
      "loss": 0.1488,
      "num_input_tokens_seen": 43739608,
      "step": 27190
    },
    {
      "epoch": 1.2311052762624777,
      "grad_norm": 0.4209730923175812,
      "learning_rate": 1.1614278616548904e-05,
      "loss": 0.1581,
      "num_input_tokens_seen": 43747256,
      "step": 27195
    },
    {
      "epoch": 1.2313316281491207,
      "grad_norm": 0.49619466066360474,
      "learning_rate": 1.1605988000843986e-05,
      "loss": 0.1313,
      "num_input_tokens_seen": 43755288,
      "step": 27200
    },
    {
      "epoch": 1.2313316281491207,
      "eval_loss": 0.14437107741832733,
      "eval_runtime": 404.6839,
      "eval_samples_per_second": 97.041,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 43755288,
      "step": 27200
    },
    {
      "epoch": 1.2315579800357637,
      "grad_norm": 0.5471482872962952,
      "learning_rate": 1.1597699450664028e-05,
      "loss": 0.1513,
      "num_input_tokens_seen": 43763320,
      "step": 27205
    },
    {
      "epoch": 1.2317843319224067,
      "grad_norm": 0.7821931838989258,
      "learning_rate": 1.1589412967287252e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 43771384,
      "step": 27210
    },
    {
      "epoch": 1.2320106838090497,
      "grad_norm": 0.7741258144378662,
      "learning_rate": 1.1581128551991514e-05,
      "loss": 0.1243,
      "num_input_tokens_seen": 43779640,
      "step": 27215
    },
    {
      "epoch": 1.2322370356956924,
      "grad_norm": 0.47402846813201904,
      "learning_rate": 1.1572846206054383e-05,
      "loss": 0.1173,
      "num_input_tokens_seen": 43787096,
      "step": 27220
    },
    {
      "epoch": 1.2324633875823354,
      "grad_norm": 0.48846328258514404,
      "learning_rate": 1.1564565930753113e-05,
      "loss": 0.1425,
      "num_input_tokens_seen": 43794808,
      "step": 27225
    },
    {
      "epoch": 1.2326897394689784,
      "grad_norm": 0.36399295926094055,
      "learning_rate": 1.1556287727364606e-05,
      "loss": 0.1187,
      "num_input_tokens_seen": 43802712,
      "step": 27230
    },
    {
      "epoch": 1.2329160913556214,
      "grad_norm": 0.47902756929397583,
      "learning_rate": 1.1548011597165489e-05,
      "loss": 0.1566,
      "num_input_tokens_seen": 43810456,
      "step": 27235
    },
    {
      "epoch": 1.2331424432422644,
      "grad_norm": 0.847976803779602,
      "learning_rate": 1.1539737541432019e-05,
      "loss": 0.1508,
      "num_input_tokens_seen": 43818456,
      "step": 27240
    },
    {
      "epoch": 1.2333687951289074,
      "grad_norm": 1.1759761571884155,
      "learning_rate": 1.1531465561440174e-05,
      "loss": 0.151,
      "num_input_tokens_seen": 43826296,
      "step": 27245
    },
    {
      "epoch": 1.2335951470155504,
      "grad_norm": 0.3840519189834595,
      "learning_rate": 1.1523195658465605e-05,
      "loss": 0.1561,
      "num_input_tokens_seen": 43834392,
      "step": 27250
    },
    {
      "epoch": 1.2338214989021934,
      "grad_norm": 0.6791654825210571,
      "learning_rate": 1.1514927833783618e-05,
      "loss": 0.114,
      "num_input_tokens_seen": 43842360,
      "step": 27255
    },
    {
      "epoch": 1.2340478507888364,
      "grad_norm": 0.7820302248001099,
      "learning_rate": 1.150666208866922e-05,
      "loss": 0.1884,
      "num_input_tokens_seen": 43849816,
      "step": 27260
    },
    {
      "epoch": 1.2342742026754794,
      "grad_norm": 0.4209316670894623,
      "learning_rate": 1.1498398424397106e-05,
      "loss": 0.1674,
      "num_input_tokens_seen": 43858008,
      "step": 27265
    },
    {
      "epoch": 1.2345005545621222,
      "grad_norm": 0.4707402288913727,
      "learning_rate": 1.1490136842241628e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 43865752,
      "step": 27270
    },
    {
      "epoch": 1.2347269064487651,
      "grad_norm": 0.7818138003349304,
      "learning_rate": 1.1481877343476813e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 43874168,
      "step": 27275
    },
    {
      "epoch": 1.2349532583354081,
      "grad_norm": 0.5100980401039124,
      "learning_rate": 1.14736199293764e-05,
      "loss": 0.161,
      "num_input_tokens_seen": 43882456,
      "step": 27280
    },
    {
      "epoch": 1.2351796102220511,
      "grad_norm": 0.8396208882331848,
      "learning_rate": 1.1465364601213771e-05,
      "loss": 0.1671,
      "num_input_tokens_seen": 43891064,
      "step": 27285
    },
    {
      "epoch": 1.2354059621086941,
      "grad_norm": 0.28856343030929565,
      "learning_rate": 1.1457111360262012e-05,
      "loss": 0.1597,
      "num_input_tokens_seen": 43898936,
      "step": 27290
    },
    {
      "epoch": 1.2356323139953371,
      "grad_norm": 0.7233585715293884,
      "learning_rate": 1.1448860207793869e-05,
      "loss": 0.1946,
      "num_input_tokens_seen": 43906808,
      "step": 27295
    },
    {
      "epoch": 1.2358586658819801,
      "grad_norm": 1.1441172361373901,
      "learning_rate": 1.144061114508177e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 43914968,
      "step": 27300
    },
    {
      "epoch": 1.2360850177686231,
      "grad_norm": 0.46094810962677,
      "learning_rate": 1.1432364173397842e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 43922872,
      "step": 27305
    },
    {
      "epoch": 1.2363113696552661,
      "grad_norm": 0.5162558555603027,
      "learning_rate": 1.1424119294013852e-05,
      "loss": 0.1545,
      "num_input_tokens_seen": 43930904,
      "step": 27310
    },
    {
      "epoch": 1.236537721541909,
      "grad_norm": 0.38824185729026794,
      "learning_rate": 1.1415876508201279e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 43938392,
      "step": 27315
    },
    {
      "epoch": 1.236764073428552,
      "grad_norm": 0.26999425888061523,
      "learning_rate": 1.140763581723125e-05,
      "loss": 0.1501,
      "num_input_tokens_seen": 43946456,
      "step": 27320
    },
    {
      "epoch": 1.236990425315195,
      "grad_norm": 0.37278619408607483,
      "learning_rate": 1.1399397222374588e-05,
      "loss": 0.1507,
      "num_input_tokens_seen": 43954040,
      "step": 27325
    },
    {
      "epoch": 1.237216777201838,
      "grad_norm": 0.90157550573349,
      "learning_rate": 1.1391160724901804e-05,
      "loss": 0.186,
      "num_input_tokens_seen": 43962232,
      "step": 27330
    },
    {
      "epoch": 1.2374431290884809,
      "grad_norm": 0.35779303312301636,
      "learning_rate": 1.138292632608304e-05,
      "loss": 0.1104,
      "num_input_tokens_seen": 43969528,
      "step": 27335
    },
    {
      "epoch": 1.2376694809751239,
      "grad_norm": 0.8725802898406982,
      "learning_rate": 1.1374694027188174e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 43977496,
      "step": 27340
    },
    {
      "epoch": 1.2378958328617669,
      "grad_norm": 0.5425011515617371,
      "learning_rate": 1.1366463829486711e-05,
      "loss": 0.163,
      "num_input_tokens_seen": 43984920,
      "step": 27345
    },
    {
      "epoch": 1.2381221847484098,
      "grad_norm": 0.6198047995567322,
      "learning_rate": 1.1358235734247849e-05,
      "loss": 0.1241,
      "num_input_tokens_seen": 43992312,
      "step": 27350
    },
    {
      "epoch": 1.2383485366350528,
      "grad_norm": 0.5623555183410645,
      "learning_rate": 1.1350009742740478e-05,
      "loss": 0.1423,
      "num_input_tokens_seen": 44001080,
      "step": 27355
    },
    {
      "epoch": 1.2385748885216958,
      "grad_norm": 0.2492474466562271,
      "learning_rate": 1.134178585623313e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 44008888,
      "step": 27360
    },
    {
      "epoch": 1.2388012404083388,
      "grad_norm": 0.8943191170692444,
      "learning_rate": 1.1333564075994047e-05,
      "loss": 0.1535,
      "num_input_tokens_seen": 44017080,
      "step": 27365
    },
    {
      "epoch": 1.2390275922949818,
      "grad_norm": 0.5886397361755371,
      "learning_rate": 1.1325344403291133e-05,
      "loss": 0.1506,
      "num_input_tokens_seen": 44024600,
      "step": 27370
    },
    {
      "epoch": 1.2392539441816248,
      "grad_norm": 0.9151039719581604,
      "learning_rate": 1.1317126839391951e-05,
      "loss": 0.1381,
      "num_input_tokens_seen": 44033048,
      "step": 27375
    },
    {
      "epoch": 1.2394802960682678,
      "grad_norm": 0.28140729665756226,
      "learning_rate": 1.1308911385563766e-05,
      "loss": 0.1623,
      "num_input_tokens_seen": 44040824,
      "step": 27380
    },
    {
      "epoch": 1.2397066479549106,
      "grad_norm": 0.6206610798835754,
      "learning_rate": 1.1300698043073494e-05,
      "loss": 0.1467,
      "num_input_tokens_seen": 44048760,
      "step": 27385
    },
    {
      "epoch": 1.2399329998415536,
      "grad_norm": 0.5115307569503784,
      "learning_rate": 1.1292486813187736e-05,
      "loss": 0.1341,
      "num_input_tokens_seen": 44056536,
      "step": 27390
    },
    {
      "epoch": 1.2401593517281966,
      "grad_norm": 0.8321195840835571,
      "learning_rate": 1.1284277697172782e-05,
      "loss": 0.1631,
      "num_input_tokens_seen": 44064568,
      "step": 27395
    },
    {
      "epoch": 1.2403857036148396,
      "grad_norm": 0.8415385484695435,
      "learning_rate": 1.127607069629456e-05,
      "loss": 0.1475,
      "num_input_tokens_seen": 44072376,
      "step": 27400
    },
    {
      "epoch": 1.2403857036148396,
      "eval_loss": 0.14484407007694244,
      "eval_runtime": 404.503,
      "eval_samples_per_second": 97.085,
      "eval_steps_per_second": 24.272,
      "num_input_tokens_seen": 44072376,
      "step": 27400
    },
    {
      "epoch": 1.2406120555014826,
      "grad_norm": 1.2576769590377808,
      "learning_rate": 1.1267865811818701e-05,
      "loss": 0.133,
      "num_input_tokens_seen": 44080600,
      "step": 27405
    },
    {
      "epoch": 1.2408384073881256,
      "grad_norm": 1.1207573413848877,
      "learning_rate": 1.1259663045010513e-05,
      "loss": 0.1092,
      "num_input_tokens_seen": 44089048,
      "step": 27410
    },
    {
      "epoch": 1.2410647592747686,
      "grad_norm": 1.0499821901321411,
      "learning_rate": 1.1251462397134957e-05,
      "loss": 0.16,
      "num_input_tokens_seen": 44097240,
      "step": 27415
    },
    {
      "epoch": 1.2412911111614116,
      "grad_norm": 0.37971678376197815,
      "learning_rate": 1.1243263869456664e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 44105752,
      "step": 27420
    },
    {
      "epoch": 1.2415174630480545,
      "grad_norm": 1.1727486848831177,
      "learning_rate": 1.1235067463239967e-05,
      "loss": 0.1565,
      "num_input_tokens_seen": 44113688,
      "step": 27425
    },
    {
      "epoch": 1.2417438149346975,
      "grad_norm": 0.6349320411682129,
      "learning_rate": 1.122687317974884e-05,
      "loss": 0.1268,
      "num_input_tokens_seen": 44121848,
      "step": 27430
    },
    {
      "epoch": 1.2419701668213405,
      "grad_norm": 0.5801634788513184,
      "learning_rate": 1.1218681020246963e-05,
      "loss": 0.1096,
      "num_input_tokens_seen": 44129720,
      "step": 27435
    },
    {
      "epoch": 1.2421965187079835,
      "grad_norm": 1.4382864236831665,
      "learning_rate": 1.1210490985997652e-05,
      "loss": 0.1655,
      "num_input_tokens_seen": 44137272,
      "step": 27440
    },
    {
      "epoch": 1.2424228705946265,
      "grad_norm": 0.5391113758087158,
      "learning_rate": 1.1202303078263917e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 44145240,
      "step": 27445
    },
    {
      "epoch": 1.2426492224812693,
      "grad_norm": 0.792692244052887,
      "learning_rate": 1.1194117298308451e-05,
      "loss": 0.1568,
      "num_input_tokens_seen": 44153176,
      "step": 27450
    },
    {
      "epoch": 1.2428755743679123,
      "grad_norm": 0.5357409715652466,
      "learning_rate": 1.1185933647393585e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 44161176,
      "step": 27455
    },
    {
      "epoch": 1.2431019262545553,
      "grad_norm": 0.5594149231910706,
      "learning_rate": 1.1177752126781354e-05,
      "loss": 0.1589,
      "num_input_tokens_seen": 44168888,
      "step": 27460
    },
    {
      "epoch": 1.2433282781411983,
      "grad_norm": 0.5156538486480713,
      "learning_rate": 1.1169572737733441e-05,
      "loss": 0.1167,
      "num_input_tokens_seen": 44177240,
      "step": 27465
    },
    {
      "epoch": 1.2435546300278413,
      "grad_norm": 0.705060601234436,
      "learning_rate": 1.1161395481511216e-05,
      "loss": 0.1738,
      "num_input_tokens_seen": 44185208,
      "step": 27470
    },
    {
      "epoch": 1.2437809819144843,
      "grad_norm": 0.7916751503944397,
      "learning_rate": 1.1153220359375722e-05,
      "loss": 0.1027,
      "num_input_tokens_seen": 44193336,
      "step": 27475
    },
    {
      "epoch": 1.2440073338011273,
      "grad_norm": 0.7070046067237854,
      "learning_rate": 1.114504737258765e-05,
      "loss": 0.1248,
      "num_input_tokens_seen": 44201880,
      "step": 27480
    },
    {
      "epoch": 1.2442336856877703,
      "grad_norm": 0.6933754682540894,
      "learning_rate": 1.1136876522407393e-05,
      "loss": 0.1311,
      "num_input_tokens_seen": 44209720,
      "step": 27485
    },
    {
      "epoch": 1.2444600375744133,
      "grad_norm": 0.9952760338783264,
      "learning_rate": 1.1128707810094985e-05,
      "loss": 0.1239,
      "num_input_tokens_seen": 44217464,
      "step": 27490
    },
    {
      "epoch": 1.2446863894610563,
      "grad_norm": 0.6995567679405212,
      "learning_rate": 1.1120541236910157e-05,
      "loss": 0.1121,
      "num_input_tokens_seen": 44225336,
      "step": 27495
    },
    {
      "epoch": 1.244912741347699,
      "grad_norm": 0.7373906373977661,
      "learning_rate": 1.111237680411229e-05,
      "loss": 0.2117,
      "num_input_tokens_seen": 44233624,
      "step": 27500
    },
    {
      "epoch": 1.245139093234342,
      "grad_norm": 0.4125010073184967,
      "learning_rate": 1.1104214512960433e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 44241304,
      "step": 27505
    },
    {
      "epoch": 1.245365445120985,
      "grad_norm": 1.2047960758209229,
      "learning_rate": 1.1096054364713327e-05,
      "loss": 0.1601,
      "num_input_tokens_seen": 44248792,
      "step": 27510
    },
    {
      "epoch": 1.245591797007628,
      "grad_norm": 0.3756188750267029,
      "learning_rate": 1.1087896360629371e-05,
      "loss": 0.1257,
      "num_input_tokens_seen": 44256856,
      "step": 27515
    },
    {
      "epoch": 1.245818148894271,
      "grad_norm": 0.49080052971839905,
      "learning_rate": 1.107974050196662e-05,
      "loss": 0.165,
      "num_input_tokens_seen": 44264792,
      "step": 27520
    },
    {
      "epoch": 1.246044500780914,
      "grad_norm": 1.0719064474105835,
      "learning_rate": 1.1071586789982816e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 44272248,
      "step": 27525
    },
    {
      "epoch": 1.246270852667557,
      "grad_norm": 1.023213267326355,
      "learning_rate": 1.1063435225935373e-05,
      "loss": 0.1586,
      "num_input_tokens_seen": 44280248,
      "step": 27530
    },
    {
      "epoch": 1.2464972045542,
      "grad_norm": 0.8995745182037354,
      "learning_rate": 1.1055285811081348e-05,
      "loss": 0.1294,
      "num_input_tokens_seen": 44288344,
      "step": 27535
    },
    {
      "epoch": 1.246723556440843,
      "grad_norm": 0.4218739867210388,
      "learning_rate": 1.1047138546677499e-05,
      "loss": 0.1477,
      "num_input_tokens_seen": 44296184,
      "step": 27540
    },
    {
      "epoch": 1.246949908327486,
      "grad_norm": 0.5452826619148254,
      "learning_rate": 1.1038993433980219e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 44304472,
      "step": 27545
    },
    {
      "epoch": 1.247176260214129,
      "grad_norm": 0.7456250786781311,
      "learning_rate": 1.1030850474245597e-05,
      "loss": 0.1809,
      "num_input_tokens_seen": 44312216,
      "step": 27550
    },
    {
      "epoch": 1.247402612100772,
      "grad_norm": 0.5038443207740784,
      "learning_rate": 1.102270966872939e-05,
      "loss": 0.1256,
      "num_input_tokens_seen": 44320056,
      "step": 27555
    },
    {
      "epoch": 1.2476289639874147,
      "grad_norm": 0.544701337814331,
      "learning_rate": 1.1014571018687e-05,
      "loss": 0.1226,
      "num_input_tokens_seen": 44327800,
      "step": 27560
    },
    {
      "epoch": 1.2478553158740577,
      "grad_norm": 0.5709578394889832,
      "learning_rate": 1.1006434525373502e-05,
      "loss": 0.1457,
      "num_input_tokens_seen": 44335896,
      "step": 27565
    },
    {
      "epoch": 1.2480816677607007,
      "grad_norm": 0.3227342367172241,
      "learning_rate": 1.0998300190043664e-05,
      "loss": 0.1517,
      "num_input_tokens_seen": 44344216,
      "step": 27570
    },
    {
      "epoch": 1.2483080196473437,
      "grad_norm": 0.9605824947357178,
      "learning_rate": 1.0990168013951882e-05,
      "loss": 0.1148,
      "num_input_tokens_seen": 44352312,
      "step": 27575
    },
    {
      "epoch": 1.2485343715339867,
      "grad_norm": 0.35641658306121826,
      "learning_rate": 1.0982037998352263e-05,
      "loss": 0.1394,
      "num_input_tokens_seen": 44360472,
      "step": 27580
    },
    {
      "epoch": 1.2487607234206297,
      "grad_norm": 0.3165603578090668,
      "learning_rate": 1.0973910144498534e-05,
      "loss": 0.131,
      "num_input_tokens_seen": 44367928,
      "step": 27585
    },
    {
      "epoch": 1.2489870753072727,
      "grad_norm": 0.40532898902893066,
      "learning_rate": 1.0965784453644123e-05,
      "loss": 0.1278,
      "num_input_tokens_seen": 44375896,
      "step": 27590
    },
    {
      "epoch": 1.2492134271939157,
      "grad_norm": 0.531857967376709,
      "learning_rate": 1.0957660927042127e-05,
      "loss": 0.1544,
      "num_input_tokens_seen": 44384344,
      "step": 27595
    },
    {
      "epoch": 1.2494397790805587,
      "grad_norm": 1.115136742591858,
      "learning_rate": 1.094953956594527e-05,
      "loss": 0.1867,
      "num_input_tokens_seen": 44391832,
      "step": 27600
    },
    {
      "epoch": 1.2494397790805587,
      "eval_loss": 0.14428825676441193,
      "eval_runtime": 403.8828,
      "eval_samples_per_second": 97.234,
      "eval_steps_per_second": 24.309,
      "num_input_tokens_seen": 44391832,
      "step": 27600
    },
    {
      "epoch": 1.2496661309672017,
      "grad_norm": 0.6394381523132324,
      "learning_rate": 1.0941420371605981e-05,
      "loss": 0.1198,
      "num_input_tokens_seen": 44399736,
      "step": 27605
    },
    {
      "epoch": 1.2498924828538445,
      "grad_norm": 0.6933061480522156,
      "learning_rate": 1.0933303345276354e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 44407096,
      "step": 27610
    },
    {
      "epoch": 1.2501188347404875,
      "grad_norm": 0.30785849690437317,
      "learning_rate": 1.0925188488208112e-05,
      "loss": 0.1062,
      "num_input_tokens_seen": 44414616,
      "step": 27615
    },
    {
      "epoch": 1.2503451866271305,
      "grad_norm": 0.4972275197505951,
      "learning_rate": 1.0917075801652694e-05,
      "loss": 0.1168,
      "num_input_tokens_seen": 44422872,
      "step": 27620
    },
    {
      "epoch": 1.2505715385137735,
      "grad_norm": 0.3509436845779419,
      "learning_rate": 1.0908965286861151e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 44430680,
      "step": 27625
    },
    {
      "epoch": 1.2507978904004164,
      "grad_norm": 0.5158130526542664,
      "learning_rate": 1.090085694508425e-05,
      "loss": 0.1377,
      "num_input_tokens_seen": 44438360,
      "step": 27630
    },
    {
      "epoch": 1.2510242422870594,
      "grad_norm": 0.7782248258590698,
      "learning_rate": 1.089275077757238e-05,
      "loss": 0.1745,
      "num_input_tokens_seen": 44446360,
      "step": 27635
    },
    {
      "epoch": 1.2512505941737024,
      "grad_norm": 0.6878868341445923,
      "learning_rate": 1.0884646785575633e-05,
      "loss": 0.1478,
      "num_input_tokens_seen": 44454488,
      "step": 27640
    },
    {
      "epoch": 1.2514769460603454,
      "grad_norm": 0.3630219101905823,
      "learning_rate": 1.0876544970343728e-05,
      "loss": 0.1223,
      "num_input_tokens_seen": 44462872,
      "step": 27645
    },
    {
      "epoch": 1.2517032979469884,
      "grad_norm": 0.5249110460281372,
      "learning_rate": 1.0868445333126082e-05,
      "loss": 0.1354,
      "num_input_tokens_seen": 44471192,
      "step": 27650
    },
    {
      "epoch": 1.2519296498336314,
      "grad_norm": 0.4942728281021118,
      "learning_rate": 1.0860347875171745e-05,
      "loss": 0.127,
      "num_input_tokens_seen": 44479544,
      "step": 27655
    },
    {
      "epoch": 1.2521560017202744,
      "grad_norm": 0.4751208424568176,
      "learning_rate": 1.0852252597729465e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 44487736,
      "step": 27660
    },
    {
      "epoch": 1.2523823536069174,
      "grad_norm": 0.35041841864585876,
      "learning_rate": 1.0844159502047615e-05,
      "loss": 0.1251,
      "num_input_tokens_seen": 44495736,
      "step": 27665
    },
    {
      "epoch": 1.2526087054935604,
      "grad_norm": 0.5362825989723206,
      "learning_rate": 1.0836068589374265e-05,
      "loss": 0.1577,
      "num_input_tokens_seen": 44503512,
      "step": 27670
    },
    {
      "epoch": 1.2528350573802034,
      "grad_norm": 0.47278496623039246,
      "learning_rate": 1.0827979860957144e-05,
      "loss": 0.1482,
      "num_input_tokens_seen": 44511992,
      "step": 27675
    },
    {
      "epoch": 1.2530614092668462,
      "grad_norm": 0.5402072668075562,
      "learning_rate": 1.0819893318043615e-05,
      "loss": 0.1306,
      "num_input_tokens_seen": 44520440,
      "step": 27680
    },
    {
      "epoch": 1.2532877611534892,
      "grad_norm": 0.65529865026474,
      "learning_rate": 1.0811808961880734e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 44528408,
      "step": 27685
    },
    {
      "epoch": 1.2535141130401322,
      "grad_norm": 0.8449394702911377,
      "learning_rate": 1.080372679371522e-05,
      "loss": 0.1097,
      "num_input_tokens_seen": 44537016,
      "step": 27690
    },
    {
      "epoch": 1.2537404649267752,
      "grad_norm": 0.7376478314399719,
      "learning_rate": 1.0795646814793428e-05,
      "loss": 0.1403,
      "num_input_tokens_seen": 44544856,
      "step": 27695
    },
    {
      "epoch": 1.2539668168134182,
      "grad_norm": 0.6916501522064209,
      "learning_rate": 1.078756902636141e-05,
      "loss": 0.1017,
      "num_input_tokens_seen": 44552824,
      "step": 27700
    },
    {
      "epoch": 1.2541931687000611,
      "grad_norm": 0.5236243009567261,
      "learning_rate": 1.077949342966485e-05,
      "loss": 0.1487,
      "num_input_tokens_seen": 44560792,
      "step": 27705
    },
    {
      "epoch": 1.2544195205867041,
      "grad_norm": 1.0505272150039673,
      "learning_rate": 1.0771420025949103e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 44569176,
      "step": 27710
    },
    {
      "epoch": 1.2546458724733471,
      "grad_norm": 0.6371818780899048,
      "learning_rate": 1.0763348816459204e-05,
      "loss": 0.132,
      "num_input_tokens_seen": 44577400,
      "step": 27715
    },
    {
      "epoch": 1.25487222435999,
      "grad_norm": 0.3335374593734741,
      "learning_rate": 1.0755279802439816e-05,
      "loss": 0.141,
      "num_input_tokens_seen": 44585176,
      "step": 27720
    },
    {
      "epoch": 1.255098576246633,
      "grad_norm": 0.5637281537055969,
      "learning_rate": 1.0747212985135293e-05,
      "loss": 0.1533,
      "num_input_tokens_seen": 44592984,
      "step": 27725
    },
    {
      "epoch": 1.255324928133276,
      "grad_norm": 0.47151198983192444,
      "learning_rate": 1.073914836578965e-05,
      "loss": 0.1659,
      "num_input_tokens_seen": 44600632,
      "step": 27730
    },
    {
      "epoch": 1.255551280019919,
      "grad_norm": 0.3513195216655731,
      "learning_rate": 1.0731085945646529e-05,
      "loss": 0.1367,
      "num_input_tokens_seen": 44608728,
      "step": 27735
    },
    {
      "epoch": 1.2557776319065619,
      "grad_norm": 0.9398630261421204,
      "learning_rate": 1.0723025725949285e-05,
      "loss": 0.1899,
      "num_input_tokens_seen": 44617080,
      "step": 27740
    },
    {
      "epoch": 1.2560039837932049,
      "grad_norm": 0.8982202410697937,
      "learning_rate": 1.0714967707940875e-05,
      "loss": 0.1765,
      "num_input_tokens_seen": 44625336,
      "step": 27745
    },
    {
      "epoch": 1.2562303356798479,
      "grad_norm": 0.4089524447917938,
      "learning_rate": 1.0706911892863963e-05,
      "loss": 0.1328,
      "num_input_tokens_seen": 44633560,
      "step": 27750
    },
    {
      "epoch": 1.2564566875664909,
      "grad_norm": 0.36071184277534485,
      "learning_rate": 1.0698858281960866e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 44641560,
      "step": 27755
    },
    {
      "epoch": 1.2566830394531339,
      "grad_norm": 0.5877086520195007,
      "learning_rate": 1.069080687647353e-05,
      "loss": 0.1067,
      "num_input_tokens_seen": 44649144,
      "step": 27760
    },
    {
      "epoch": 1.2569093913397769,
      "grad_norm": 0.7373748421669006,
      "learning_rate": 1.0682757677643596e-05,
      "loss": 0.1756,
      "num_input_tokens_seen": 44657272,
      "step": 27765
    },
    {
      "epoch": 1.2571357432264199,
      "grad_norm": 0.5900368094444275,
      "learning_rate": 1.0674710686712359e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 44665240,
      "step": 27770
    },
    {
      "epoch": 1.2573620951130628,
      "grad_norm": 0.7571877241134644,
      "learning_rate": 1.0666665904920756e-05,
      "loss": 0.1193,
      "num_input_tokens_seen": 44673592,
      "step": 27775
    },
    {
      "epoch": 1.2575884469997058,
      "grad_norm": 0.44814106822013855,
      "learning_rate": 1.0658623333509385e-05,
      "loss": 0.1305,
      "num_input_tokens_seen": 44681176,
      "step": 27780
    },
    {
      "epoch": 1.2578147988863488,
      "grad_norm": 0.8639662265777588,
      "learning_rate": 1.0650582973718532e-05,
      "loss": 0.1547,
      "num_input_tokens_seen": 44689816,
      "step": 27785
    },
    {
      "epoch": 1.2580411507729916,
      "grad_norm": 0.9918429255485535,
      "learning_rate": 1.0642544826788098e-05,
      "loss": 0.1163,
      "num_input_tokens_seen": 44697944,
      "step": 27790
    },
    {
      "epoch": 1.2582675026596346,
      "grad_norm": 0.7538689970970154,
      "learning_rate": 1.063450889395769e-05,
      "loss": 0.1154,
      "num_input_tokens_seen": 44706232,
      "step": 27795
    },
    {
      "epoch": 1.2584938545462776,
      "grad_norm": 0.465408056974411,
      "learning_rate": 1.062647517646653e-05,
      "loss": 0.1318,
      "num_input_tokens_seen": 44714776,
      "step": 27800
    },
    {
      "epoch": 1.2584938545462776,
      "eval_loss": 0.14480425417423248,
      "eval_runtime": 405.2217,
      "eval_samples_per_second": 96.912,
      "eval_steps_per_second": 24.229,
      "num_input_tokens_seen": 44714776,
      "step": 27800
    },
    {
      "epoch": 1.2587202064329206,
      "grad_norm": 0.677021324634552,
      "learning_rate": 1.0618443675553527e-05,
      "loss": 0.1146,
      "num_input_tokens_seen": 44723096,
      "step": 27805
    },
    {
      "epoch": 1.2589465583195636,
      "grad_norm": 0.5944973826408386,
      "learning_rate": 1.0610414392457247e-05,
      "loss": 0.1621,
      "num_input_tokens_seen": 44731096,
      "step": 27810
    },
    {
      "epoch": 1.2591729102062066,
      "grad_norm": 0.93985915184021,
      "learning_rate": 1.0602387328415888e-05,
      "loss": 0.1342,
      "num_input_tokens_seen": 44738616,
      "step": 27815
    },
    {
      "epoch": 1.2593992620928496,
      "grad_norm": 0.8400663137435913,
      "learning_rate": 1.0594362484667347e-05,
      "loss": 0.1156,
      "num_input_tokens_seen": 44746424,
      "step": 27820
    },
    {
      "epoch": 1.2596256139794926,
      "grad_norm": 1.0183305740356445,
      "learning_rate": 1.0586339862449132e-05,
      "loss": 0.1315,
      "num_input_tokens_seen": 44754456,
      "step": 27825
    },
    {
      "epoch": 1.2598519658661356,
      "grad_norm": 0.6057889461517334,
      "learning_rate": 1.0578319462998445e-05,
      "loss": 0.1489,
      "num_input_tokens_seen": 44762776,
      "step": 27830
    },
    {
      "epoch": 1.2600783177527783,
      "grad_norm": 0.6114991307258606,
      "learning_rate": 1.057030128755214e-05,
      "loss": 0.1385,
      "num_input_tokens_seen": 44770584,
      "step": 27835
    },
    {
      "epoch": 1.2603046696394213,
      "grad_norm": 0.6875095367431641,
      "learning_rate": 1.0562285337346703e-05,
      "loss": 0.1376,
      "num_input_tokens_seen": 44778936,
      "step": 27840
    },
    {
      "epoch": 1.2605310215260643,
      "grad_norm": 0.5365665555000305,
      "learning_rate": 1.0554271613618308e-05,
      "loss": 0.1612,
      "num_input_tokens_seen": 44786936,
      "step": 27845
    },
    {
      "epoch": 1.2607573734127073,
      "grad_norm": 0.5593551397323608,
      "learning_rate": 1.054626011760276e-05,
      "loss": 0.1415,
      "num_input_tokens_seen": 44795448,
      "step": 27850
    },
    {
      "epoch": 1.2609837252993503,
      "grad_norm": 0.7110141515731812,
      "learning_rate": 1.0538250850535549e-05,
      "loss": 0.1382,
      "num_input_tokens_seen": 44803544,
      "step": 27855
    },
    {
      "epoch": 1.2612100771859933,
      "grad_norm": 0.8046485185623169,
      "learning_rate": 1.0530243813651794e-05,
      "loss": 0.1452,
      "num_input_tokens_seen": 44811352,
      "step": 27860
    },
    {
      "epoch": 1.2614364290726363,
      "grad_norm": 0.4751913249492645,
      "learning_rate": 1.0522239008186271e-05,
      "loss": 0.1272,
      "num_input_tokens_seen": 44819448,
      "step": 27865
    },
    {
      "epoch": 1.2616627809592793,
      "grad_norm": 0.5806528925895691,
      "learning_rate": 1.0514236435373434e-05,
      "loss": 0.1121,
      "num_input_tokens_seen": 44827224,
      "step": 27870
    },
    {
      "epoch": 1.2618891328459223,
      "grad_norm": 0.5766690969467163,
      "learning_rate": 1.0506236096447386e-05,
      "loss": 0.1584,
      "num_input_tokens_seen": 44835768,
      "step": 27875
    },
    {
      "epoch": 1.2621154847325653,
      "grad_norm": 0.5663761496543884,
      "learning_rate": 1.049823799264186e-05,
      "loss": 0.1587,
      "num_input_tokens_seen": 44843512,
      "step": 27880
    },
    {
      "epoch": 1.2623418366192083,
      "grad_norm": 1.1669193506240845,
      "learning_rate": 1.049024212519028e-05,
      "loss": 0.2001,
      "num_input_tokens_seen": 44851960,
      "step": 27885
    },
    {
      "epoch": 1.2625681885058513,
      "grad_norm": 0.5099561214447021,
      "learning_rate": 1.0482248495325713e-05,
      "loss": 0.1182,
      "num_input_tokens_seen": 44859864,
      "step": 27890
    },
    {
      "epoch": 1.2627945403924943,
      "grad_norm": 0.441707044839859,
      "learning_rate": 1.047425710428086e-05,
      "loss": 0.1714,
      "num_input_tokens_seen": 44867736,
      "step": 27895
    },
    {
      "epoch": 1.2630208922791373,
      "grad_norm": 0.424360990524292,
      "learning_rate": 1.0466267953288114e-05,
      "loss": 0.1338,
      "num_input_tokens_seen": 44875256,
      "step": 27900
    },
    {
      "epoch": 1.26324724416578,
      "grad_norm": 1.3623446226119995,
      "learning_rate": 1.0458281043579482e-05,
      "loss": 0.1326,
      "num_input_tokens_seen": 44883096,
      "step": 27905
    },
    {
      "epoch": 1.263473596052423,
      "grad_norm": 0.904360294342041,
      "learning_rate": 1.0450296376386657e-05,
      "loss": 0.143,
      "num_input_tokens_seen": 44890936,
      "step": 27910
    },
    {
      "epoch": 1.263699947939066,
      "grad_norm": 0.43439412117004395,
      "learning_rate": 1.044231395294098e-05,
      "loss": 0.1549,
      "num_input_tokens_seen": 44899064,
      "step": 27915
    },
    {
      "epoch": 1.263926299825709,
      "grad_norm": 0.5579603314399719,
      "learning_rate": 1.0434333774473435e-05,
      "loss": 0.1235,
      "num_input_tokens_seen": 44907064,
      "step": 27920
    },
    {
      "epoch": 1.264152651712352,
      "grad_norm": 0.9926174283027649,
      "learning_rate": 1.0426355842214657e-05,
      "loss": 0.1349,
      "num_input_tokens_seen": 44915544,
      "step": 27925
    },
    {
      "epoch": 1.264379003598995,
      "grad_norm": 0.4042283594608307,
      "learning_rate": 1.0418380157394963e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 44922744,
      "step": 27930
    },
    {
      "epoch": 1.264605355485638,
      "grad_norm": 0.7925674319267273,
      "learning_rate": 1.0410406721244281e-05,
      "loss": 0.1617,
      "num_input_tokens_seen": 44930680,
      "step": 27935
    },
    {
      "epoch": 1.264831707372281,
      "grad_norm": 0.8926230072975159,
      "learning_rate": 1.0402435534992238e-05,
      "loss": 0.1448,
      "num_input_tokens_seen": 44938776,
      "step": 27940
    },
    {
      "epoch": 1.265058059258924,
      "grad_norm": 0.31652718782424927,
      "learning_rate": 1.0394466599868071e-05,
      "loss": 0.1136,
      "num_input_tokens_seen": 44946744,
      "step": 27945
    },
    {
      "epoch": 1.2652844111455668,
      "grad_norm": 0.3839770555496216,
      "learning_rate": 1.0386499917100697e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 44954104,
      "step": 27950
    },
    {
      "epoch": 1.2655107630322098,
      "grad_norm": 1.270076036453247,
      "learning_rate": 1.0378535487918692e-05,
      "loss": 0.1502,
      "num_input_tokens_seen": 44961848,
      "step": 27955
    },
    {
      "epoch": 1.2657371149188528,
      "grad_norm": 0.6076760292053223,
      "learning_rate": 1.037057331355025e-05,
      "loss": 0.1724,
      "num_input_tokens_seen": 44969432,
      "step": 27960
    },
    {
      "epoch": 1.2659634668054958,
      "grad_norm": 0.9091634750366211,
      "learning_rate": 1.0362613395223247e-05,
      "loss": 0.1416,
      "num_input_tokens_seen": 44977944,
      "step": 27965
    },
    {
      "epoch": 1.2661898186921388,
      "grad_norm": 0.7234622836112976,
      "learning_rate": 1.0354655734165212e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 44986328,
      "step": 27970
    },
    {
      "epoch": 1.2664161705787818,
      "grad_norm": 0.48472487926483154,
      "learning_rate": 1.03467003316033e-05,
      "loss": 0.1436,
      "num_input_tokens_seen": 44994424,
      "step": 27975
    },
    {
      "epoch": 1.2666425224654247,
      "grad_norm": 0.6480870246887207,
      "learning_rate": 1.033874718876435e-05,
      "loss": 0.145,
      "num_input_tokens_seen": 45002136,
      "step": 27980
    },
    {
      "epoch": 1.2668688743520677,
      "grad_norm": 1.0251518487930298,
      "learning_rate": 1.0330796306874818e-05,
      "loss": 0.1526,
      "num_input_tokens_seen": 45010008,
      "step": 27985
    },
    {
      "epoch": 1.2670952262387107,
      "grad_norm": 0.371282160282135,
      "learning_rate": 1.032284768716085e-05,
      "loss": 0.1279,
      "num_input_tokens_seen": 45018680,
      "step": 27990
    },
    {
      "epoch": 1.2673215781253537,
      "grad_norm": 0.3564281165599823,
      "learning_rate": 1.0314901330848206e-05,
      "loss": 0.1426,
      "num_input_tokens_seen": 45026840,
      "step": 27995
    },
    {
      "epoch": 1.2675479300119967,
      "grad_norm": 0.2662351429462433,
      "learning_rate": 1.030695723916233e-05,
      "loss": 0.138,
      "num_input_tokens_seen": 45035480,
      "step": 28000
    },
    {
      "epoch": 1.2675479300119967,
      "eval_loss": 0.14489680528640747,
      "eval_runtime": 404.951,
      "eval_samples_per_second": 96.977,
      "eval_steps_per_second": 24.245,
      "num_input_tokens_seen": 45035480,
      "step": 28000
    },
    {
      "epoch": 1.2677742818986397,
      "grad_norm": 1.0939048528671265,
      "learning_rate": 1.0299015413328289e-05,
      "loss": 0.1345,
      "num_input_tokens_seen": 45043576,
      "step": 28005
    },
    {
      "epoch": 1.2680006337852827,
      "grad_norm": 0.6997345685958862,
      "learning_rate": 1.0291075854570809e-05,
      "loss": 0.1131,
      "num_input_tokens_seen": 45051672,
      "step": 28010
    },
    {
      "epoch": 1.2682269856719257,
      "grad_norm": 0.8782492280006409,
      "learning_rate": 1.0283138564114275e-05,
      "loss": 0.1086,
      "num_input_tokens_seen": 45059448,
      "step": 28015
    },
    {
      "epoch": 1.2684533375585685,
      "grad_norm": 1.2199687957763672,
      "learning_rate": 1.027520354318273e-05,
      "loss": 0.0962,
      "num_input_tokens_seen": 45067512,
      "step": 28020
    },
    {
      "epoch": 1.2686796894452115,
      "grad_norm": 0.601256787776947,
      "learning_rate": 1.0267270792999828e-05,
      "loss": 0.1459,
      "num_input_tokens_seen": 45075640,
      "step": 28025
    },
    {
      "epoch": 1.2689060413318545,
      "grad_norm": 1.1703492403030396,
      "learning_rate": 1.0259340314788919e-05,
      "loss": 0.1054,
      "num_input_tokens_seen": 45083672,
      "step": 28030
    },
    {
      "epoch": 1.2691323932184975,
      "grad_norm": 0.3356333374977112,
      "learning_rate": 1.0251412109772979e-05,
      "loss": 0.1437,
      "num_input_tokens_seen": 45091256,
      "step": 28035
    },
    {
      "epoch": 1.2693587451051405,
      "grad_norm": 0.5497176051139832,
      "learning_rate": 1.0243486179174627e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 45099576,
      "step": 28040
    },
    {
      "epoch": 1.2695850969917835,
      "grad_norm": 0.6745988130569458,
      "learning_rate": 1.0235562524216158e-05,
      "loss": 0.1643,
      "num_input_tokens_seen": 45107096,
      "step": 28045
    },
    {
      "epoch": 1.2698114488784265,
      "grad_norm": 0.9073225855827332,
      "learning_rate": 1.022764114611948e-05,
      "loss": 0.1591,
      "num_input_tokens_seen": 45115128,
      "step": 28050
    },
    {
      "epoch": 1.2700378007650694,
      "grad_norm": 0.8330231308937073,
      "learning_rate": 1.0219722046106178e-05,
      "loss": 0.1649,
      "num_input_tokens_seen": 45123512,
      "step": 28055
    },
    {
      "epoch": 1.2702641526517122,
      "grad_norm": 0.3678068518638611,
      "learning_rate": 1.0211805225397486e-05,
      "loss": 0.1052,
      "num_input_tokens_seen": 45131736,
      "step": 28060
    },
    {
      "epoch": 1.2704905045383552,
      "grad_norm": 0.7955629229545593,
      "learning_rate": 1.020389068521426e-05,
      "loss": 0.1296,
      "num_input_tokens_seen": 45138776,
      "step": 28065
    },
    {
      "epoch": 1.2707168564249982,
      "grad_norm": 0.6713808178901672,
      "learning_rate": 1.0195978426777039e-05,
      "loss": 0.1473,
      "num_input_tokens_seen": 45146136,
      "step": 28070
    },
    {
      "epoch": 1.2709432083116412,
      "grad_norm": 0.7738249897956848,
      "learning_rate": 1.0188068451305982e-05,
      "loss": 0.2022,
      "num_input_tokens_seen": 45153848,
      "step": 28075
    },
    {
      "epoch": 1.2711695601982842,
      "grad_norm": 0.502420961856842,
      "learning_rate": 1.0180160760020902e-05,
      "loss": 0.1194,
      "num_input_tokens_seen": 45162840,
      "step": 28080
    },
    {
      "epoch": 1.2713959120849272,
      "grad_norm": 0.5338340997695923,
      "learning_rate": 1.0172255354141278e-05,
      "loss": 0.2068,
      "num_input_tokens_seen": 45170200,
      "step": 28085
    },
    {
      "epoch": 1.2716222639715702,
      "grad_norm": 0.39536428451538086,
      "learning_rate": 1.0164352234886205e-05,
      "loss": 0.1525,
      "num_input_tokens_seen": 45178040,
      "step": 28090
    },
    {
      "epoch": 1.2718486158582132,
      "grad_norm": 0.770623505115509,
      "learning_rate": 1.0156451403474454e-05,
      "loss": 0.1282,
      "num_input_tokens_seen": 45185656,
      "step": 28095
    },
    {
      "epoch": 1.2720749677448562,
      "grad_norm": 0.9172327518463135,
      "learning_rate": 1.0148552861124443e-05,
      "loss": 0.1324,
      "num_input_tokens_seen": 45193752,
      "step": 28100
    },
    {
      "epoch": 1.2723013196314992,
      "grad_norm": 1.334067702293396,
      "learning_rate": 1.0140656609054205e-05,
      "loss": 0.1471,
      "num_input_tokens_seen": 45202104,
      "step": 28105
    },
    {
      "epoch": 1.2725276715181422,
      "grad_norm": 0.6274918913841248,
      "learning_rate": 1.0132762648481455e-05,
      "loss": 0.1397,
      "num_input_tokens_seen": 45210296,
      "step": 28110
    },
    {
      "epoch": 1.2727540234047852,
      "grad_norm": 0.4405425488948822,
      "learning_rate": 1.0124870980623543e-05,
      "loss": 0.1103,
      "num_input_tokens_seen": 45218008,
      "step": 28115
    },
    {
      "epoch": 1.2729803752914282,
      "grad_norm": 0.5577871203422546,
      "learning_rate": 1.0116981606697453e-05,
      "loss": 0.156,
      "num_input_tokens_seen": 45226328,
      "step": 28120
    },
    {
      "epoch": 1.2732067271780712,
      "grad_norm": 0.4756196141242981,
      "learning_rate": 1.0109094527919838e-05,
      "loss": 0.1332,
      "num_input_tokens_seen": 45234264,
      "step": 28125
    },
    {
      "epoch": 1.273433079064714,
      "grad_norm": 0.4199269413948059,
      "learning_rate": 1.010120974550697e-05,
      "loss": 0.1207,
      "num_input_tokens_seen": 45242712,
      "step": 28130
    },
    {
      "epoch": 1.273659430951357,
      "grad_norm": 0.535710871219635,
      "learning_rate": 1.0093327260674795e-05,
      "loss": 0.149,
      "num_input_tokens_seen": 45251480,
      "step": 28135
    },
    {
      "epoch": 1.273885782838,
      "grad_norm": 0.22007405757904053,
      "learning_rate": 1.0085447074638878e-05,
      "loss": 0.1112,
      "num_input_tokens_seen": 45259480,
      "step": 28140
    },
    {
      "epoch": 1.274112134724643,
      "grad_norm": 0.9560530781745911,
      "learning_rate": 1.0077569188614461e-05,
      "loss": 0.1412,
      "num_input_tokens_seen": 45267544,
      "step": 28145
    },
    {
      "epoch": 1.274338486611286,
      "grad_norm": 0.9866092801094055,
      "learning_rate": 1.0069693603816393e-05,
      "loss": 0.147,
      "num_input_tokens_seen": 45275960,
      "step": 28150
    },
    {
      "epoch": 1.274564838497929,
      "grad_norm": 1.2736091613769531,
      "learning_rate": 1.0061820321459204e-05,
      "loss": 0.1293,
      "num_input_tokens_seen": 45283992,
      "step": 28155
    },
    {
      "epoch": 1.274791190384572,
      "grad_norm": 0.6971456408500671,
      "learning_rate": 1.0053949342757038e-05,
      "loss": 0.136,
      "num_input_tokens_seen": 45292152,
      "step": 28160
    },
    {
      "epoch": 1.2750175422712149,
      "grad_norm": 0.2489110231399536,
      "learning_rate": 1.0046080668923717e-05,
      "loss": 0.103,
      "num_input_tokens_seen": 45300280,
      "step": 28165
    },
    {
      "epoch": 1.2752438941578579,
      "grad_norm": 0.5712100863456726,
      "learning_rate": 1.003821430117267e-05,
      "loss": 0.1828,
      "num_input_tokens_seen": 45308152,
      "step": 28170
    },
    {
      "epoch": 1.2754702460445007,
      "grad_norm": 0.7622741460800171,
      "learning_rate": 1.0030350240716999e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 45315896,
      "step": 28175
    },
    {
      "epoch": 1.2756965979311436,
      "grad_norm": 0.6427751779556274,
      "learning_rate": 1.0022488488769449e-05,
      "loss": 0.1395,
      "num_input_tokens_seen": 45323736,
      "step": 28180
    },
    {
      "epoch": 1.2759229498177866,
      "grad_norm": 1.5694811344146729,
      "learning_rate": 1.0014629046542387e-05,
      "loss": 0.1493,
      "num_input_tokens_seen": 45332248,
      "step": 28185
    },
    {
      "epoch": 1.2761493017044296,
      "grad_norm": 0.41564881801605225,
      "learning_rate": 1.0006771915247842e-05,
      "loss": 0.1456,
      "num_input_tokens_seen": 45339832,
      "step": 28190
    },
    {
      "epoch": 1.2763756535910726,
      "grad_norm": 0.49394723773002625,
      "learning_rate": 9.998917096097495e-06,
      "loss": 0.1278,
      "num_input_tokens_seen": 45347256,
      "step": 28195
    },
    {
      "epoch": 1.2766020054777156,
      "grad_norm": 0.9136525392532349,
      "learning_rate": 9.991064590302638e-06,
      "loss": 0.1451,
      "num_input_tokens_seen": 45354808,
      "step": 28200
    },
    {
      "epoch": 1.2766020054777156,
      "eval_loss": 0.1441708356142044,
      "eval_runtime": 405.4173,
      "eval_samples_per_second": 96.866,
      "eval_steps_per_second": 24.217,
      "num_input_tokens_seen": 45354808,
      "step": 28200
    },
    {
      "epoch": 1.2768283573643586,
      "grad_norm": 0.7375271916389465,
      "learning_rate": 9.983214399074241e-06,
      "loss": 0.1448,
      "num_input_tokens_seen": 45363032,
      "step": 28205
    },
    {
      "epoch": 1.2770547092510016,
      "grad_norm": 0.3485172986984253,
      "learning_rate": 9.975366523622893e-06,
      "loss": 0.1653,
      "num_input_tokens_seen": 45371576,
      "step": 28210
    },
    {
      "epoch": 1.2772810611376446,
      "grad_norm": 0.3910788893699646,
      "learning_rate": 9.967520965158841e-06,
      "loss": 0.1186,
      "num_input_tokens_seen": 45379928,
      "step": 28215
    },
    {
      "epoch": 1.2775074130242876,
      "grad_norm": 0.48988932371139526,
      "learning_rate": 9.95967772489197e-06,
      "loss": 0.1213,
      "num_input_tokens_seen": 45387928,
      "step": 28220
    },
    {
      "epoch": 1.2777337649109306,
      "grad_norm": 0.6749916672706604,
      "learning_rate": 9.951836804031794e-06,
      "loss": 0.1134,
      "num_input_tokens_seen": 45396312,
      "step": 28225
    },
    {
      "epoch": 1.2779601167975736,
      "grad_norm": 0.4157683253288269,
      "learning_rate": 9.943998203787489e-06,
      "loss": 0.1634,
      "num_input_tokens_seen": 45404376,
      "step": 28230
    },
    {
      "epoch": 1.2781864686842166,
      "grad_norm": 0.8568686246871948,
      "learning_rate": 9.936161925367874e-06,
      "loss": 0.1135,
      "num_input_tokens_seen": 45412472,
      "step": 28235
    },
    {
      "epoch": 1.2784128205708596,
      "grad_norm": 0.6868199706077576,
      "learning_rate": 9.928327969981386e-06,
      "loss": 0.1364,
      "num_input_tokens_seen": 45419896,
      "step": 28240
    },
    {
      "epoch": 1.2786391724575024,
      "grad_norm": 0.9252520799636841,
      "learning_rate": 9.920496338836135e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 45428344,
      "step": 28245
    },
    {
      "epoch": 1.2788655243441454,
      "grad_norm": 1.1367871761322021,
      "learning_rate": 9.912667033139844e-06,
      "loss": 0.1439,
      "num_input_tokens_seen": 45436696,
      "step": 28250
    },
    {
      "epoch": 1.2790918762307883,
      "grad_norm": 0.9962135553359985,
      "learning_rate": 9.904840054099893e-06,
      "loss": 0.1847,
      "num_input_tokens_seen": 45444792,
      "step": 28255
    },
    {
      "epoch": 1.2793182281174313,
      "grad_norm": 0.8954480886459351,
      "learning_rate": 9.897015402923312e-06,
      "loss": 0.1561,
      "num_input_tokens_seen": 45452824,
      "step": 28260
    },
    {
      "epoch": 1.2795445800040743,
      "grad_norm": 0.6837625503540039,
      "learning_rate": 9.889193080816744e-06,
      "loss": 0.1073,
      "num_input_tokens_seen": 45460600,
      "step": 28265
    },
    {
      "epoch": 1.2797709318907173,
      "grad_norm": 0.3917417526245117,
      "learning_rate": 9.881373088986498e-06,
      "loss": 0.123,
      "num_input_tokens_seen": 45468024,
      "step": 28270
    },
    {
      "epoch": 1.2799972837773603,
      "grad_norm": 0.7423473596572876,
      "learning_rate": 9.873555428638523e-06,
      "loss": 0.1587,
      "num_input_tokens_seen": 45475512,
      "step": 28275
    },
    {
      "epoch": 1.2802236356640033,
      "grad_norm": 0.5040386319160461,
      "learning_rate": 9.865740100978383e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 45484184,
      "step": 28280
    },
    {
      "epoch": 1.280449987550646,
      "grad_norm": 0.733131468296051,
      "learning_rate": 9.857927107211315e-06,
      "loss": 0.1277,
      "num_input_tokens_seen": 45492184,
      "step": 28285
    },
    {
      "epoch": 1.280676339437289,
      "grad_norm": 0.8239191770553589,
      "learning_rate": 9.850116448542177e-06,
      "loss": 0.1704,
      "num_input_tokens_seen": 45500280,
      "step": 28290
    },
    {
      "epoch": 1.280902691323932,
      "grad_norm": 0.8800064921379089,
      "learning_rate": 9.842308126175457e-06,
      "loss": 0.162,
      "num_input_tokens_seen": 45508408,
      "step": 28295
    },
    {
      "epoch": 1.281129043210575,
      "grad_norm": 0.4188280999660492,
      "learning_rate": 9.834502141315315e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 45516056,
      "step": 28300
    },
    {
      "epoch": 1.281355395097218,
      "grad_norm": 0.4514196813106537,
      "learning_rate": 9.82669849516552e-06,
      "loss": 0.1765,
      "num_input_tokens_seen": 45523736,
      "step": 28305
    },
    {
      "epoch": 1.281581746983861,
      "grad_norm": 0.45043250918388367,
      "learning_rate": 9.818897188929493e-06,
      "loss": 0.1125,
      "num_input_tokens_seen": 45532088,
      "step": 28310
    },
    {
      "epoch": 1.281808098870504,
      "grad_norm": 1.2114708423614502,
      "learning_rate": 9.811098223810309e-06,
      "loss": 0.173,
      "num_input_tokens_seen": 45539832,
      "step": 28315
    },
    {
      "epoch": 1.282034450757147,
      "grad_norm": 0.5116134881973267,
      "learning_rate": 9.803301601010641e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 45548024,
      "step": 28320
    },
    {
      "epoch": 1.28226080264379,
      "grad_norm": 0.9556857347488403,
      "learning_rate": 9.795507321732853e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 45556408,
      "step": 28325
    },
    {
      "epoch": 1.282487154530433,
      "grad_norm": 0.663539707660675,
      "learning_rate": 9.787715387178898e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 45564536,
      "step": 28330
    },
    {
      "epoch": 1.282713506417076,
      "grad_norm": 0.7037435173988342,
      "learning_rate": 9.779925798550399e-06,
      "loss": 0.1549,
      "num_input_tokens_seen": 45572536,
      "step": 28335
    },
    {
      "epoch": 1.282939858303719,
      "grad_norm": 0.8136950731277466,
      "learning_rate": 9.772138557048619e-06,
      "loss": 0.1545,
      "num_input_tokens_seen": 45580760,
      "step": 28340
    },
    {
      "epoch": 1.283166210190362,
      "grad_norm": 0.6841100454330444,
      "learning_rate": 9.764353663874426e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 45588504,
      "step": 28345
    },
    {
      "epoch": 1.283392562077005,
      "grad_norm": 0.6523277163505554,
      "learning_rate": 9.756571120228375e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 45596696,
      "step": 28350
    },
    {
      "epoch": 1.283618913963648,
      "grad_norm": 0.9937631487846375,
      "learning_rate": 9.748790927310605e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 45605208,
      "step": 28355
    },
    {
      "epoch": 1.2838452658502908,
      "grad_norm": 1.1183905601501465,
      "learning_rate": 9.741013086320946e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 45613272,
      "step": 28360
    },
    {
      "epoch": 1.2840716177369338,
      "grad_norm": 0.6795423030853271,
      "learning_rate": 9.733237598458821e-06,
      "loss": 0.1287,
      "num_input_tokens_seen": 45621048,
      "step": 28365
    },
    {
      "epoch": 1.2842979696235768,
      "grad_norm": 0.3262880742549896,
      "learning_rate": 9.725464464923308e-06,
      "loss": 0.15,
      "num_input_tokens_seen": 45629112,
      "step": 28370
    },
    {
      "epoch": 1.2845243215102198,
      "grad_norm": 0.4800831079483032,
      "learning_rate": 9.717693686913123e-06,
      "loss": 0.1256,
      "num_input_tokens_seen": 45637368,
      "step": 28375
    },
    {
      "epoch": 1.2847506733968628,
      "grad_norm": 0.8932653665542603,
      "learning_rate": 9.709925265626632e-06,
      "loss": 0.1648,
      "num_input_tokens_seen": 45645720,
      "step": 28380
    },
    {
      "epoch": 1.2849770252835058,
      "grad_norm": 0.9144971370697021,
      "learning_rate": 9.702159202261801e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 45653528,
      "step": 28385
    },
    {
      "epoch": 1.2852033771701488,
      "grad_norm": 0.6843105554580688,
      "learning_rate": 9.694395498016268e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 45661240,
      "step": 28390
    },
    {
      "epoch": 1.2854297290567918,
      "grad_norm": 1.0435491800308228,
      "learning_rate": 9.686634154087298e-06,
      "loss": 0.1573,
      "num_input_tokens_seen": 45670616,
      "step": 28395
    },
    {
      "epoch": 1.2856560809434345,
      "grad_norm": 0.7659091949462891,
      "learning_rate": 9.678875171671776e-06,
      "loss": 0.1608,
      "num_input_tokens_seen": 45678200,
      "step": 28400
    },
    {
      "epoch": 1.2856560809434345,
      "eval_loss": 0.1440698206424713,
      "eval_runtime": 405.2191,
      "eval_samples_per_second": 96.913,
      "eval_steps_per_second": 24.229,
      "num_input_tokens_seen": 45678200,
      "step": 28400
    },
    {
      "epoch": 1.2858824328300775,
      "grad_norm": 0.3476698100566864,
      "learning_rate": 9.671118551966246e-06,
      "loss": 0.1253,
      "num_input_tokens_seen": 45686360,
      "step": 28405
    },
    {
      "epoch": 1.2861087847167205,
      "grad_norm": 0.775825560092926,
      "learning_rate": 9.66336429616686e-06,
      "loss": 0.1147,
      "num_input_tokens_seen": 45694776,
      "step": 28410
    },
    {
      "epoch": 1.2863351366033635,
      "grad_norm": 0.3289269804954529,
      "learning_rate": 9.655612405469436e-06,
      "loss": 0.1312,
      "num_input_tokens_seen": 45703128,
      "step": 28415
    },
    {
      "epoch": 1.2865614884900065,
      "grad_norm": 0.8025235533714294,
      "learning_rate": 9.647862881069413e-06,
      "loss": 0.1697,
      "num_input_tokens_seen": 45710904,
      "step": 28420
    },
    {
      "epoch": 1.2867878403766495,
      "grad_norm": 0.44711023569107056,
      "learning_rate": 9.640115724161855e-06,
      "loss": 0.1118,
      "num_input_tokens_seen": 45718680,
      "step": 28425
    },
    {
      "epoch": 1.2870141922632925,
      "grad_norm": 0.5219710469245911,
      "learning_rate": 9.632370935941483e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 45726552,
      "step": 28430
    },
    {
      "epoch": 1.2872405441499355,
      "grad_norm": 0.46714305877685547,
      "learning_rate": 9.624628517602634e-06,
      "loss": 0.1098,
      "num_input_tokens_seen": 45734584,
      "step": 28435
    },
    {
      "epoch": 1.2874668960365785,
      "grad_norm": 0.46375030279159546,
      "learning_rate": 9.61688847033928e-06,
      "loss": 0.1123,
      "num_input_tokens_seen": 45742968,
      "step": 28440
    },
    {
      "epoch": 1.2876932479232215,
      "grad_norm": 0.6719965934753418,
      "learning_rate": 9.609150795345051e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 45750584,
      "step": 28445
    },
    {
      "epoch": 1.2879195998098645,
      "grad_norm": 0.8871573209762573,
      "learning_rate": 9.601415493813171e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 45759256,
      "step": 28450
    },
    {
      "epoch": 1.2881459516965075,
      "grad_norm": 0.4022783637046814,
      "learning_rate": 9.593682566936533e-06,
      "loss": 0.1882,
      "num_input_tokens_seen": 45767160,
      "step": 28455
    },
    {
      "epoch": 1.2883723035831505,
      "grad_norm": 0.6070464253425598,
      "learning_rate": 9.58595201590766e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 45774968,
      "step": 28460
    },
    {
      "epoch": 1.2885986554697935,
      "grad_norm": 1.067456603050232,
      "learning_rate": 9.578223841918681e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 45782840,
      "step": 28465
    },
    {
      "epoch": 1.2888250073564362,
      "grad_norm": 0.6101238131523132,
      "learning_rate": 9.570498046161389e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 45790616,
      "step": 28470
    },
    {
      "epoch": 1.2890513592430792,
      "grad_norm": 0.963213324546814,
      "learning_rate": 9.562774629827206e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 45798328,
      "step": 28475
    },
    {
      "epoch": 1.2892777111297222,
      "grad_norm": 0.6449623703956604,
      "learning_rate": 9.555053594107163e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 45806200,
      "step": 28480
    },
    {
      "epoch": 1.2895040630163652,
      "grad_norm": 0.7138771414756775,
      "learning_rate": 9.547334940191957e-06,
      "loss": 0.1738,
      "num_input_tokens_seen": 45814680,
      "step": 28485
    },
    {
      "epoch": 1.2897304149030082,
      "grad_norm": 0.5079030990600586,
      "learning_rate": 9.539618669271886e-06,
      "loss": 0.103,
      "num_input_tokens_seen": 45822488,
      "step": 28490
    },
    {
      "epoch": 1.2899567667896512,
      "grad_norm": 0.7426104545593262,
      "learning_rate": 9.531904782536904e-06,
      "loss": 0.1572,
      "num_input_tokens_seen": 45830616,
      "step": 28495
    },
    {
      "epoch": 1.2901831186762942,
      "grad_norm": 0.5998918414115906,
      "learning_rate": 9.524193281176597e-06,
      "loss": 0.1479,
      "num_input_tokens_seen": 45838712,
      "step": 28500
    },
    {
      "epoch": 1.2904094705629372,
      "grad_norm": 0.4254026710987091,
      "learning_rate": 9.516484166380165e-06,
      "loss": 0.1296,
      "num_input_tokens_seen": 45846776,
      "step": 28505
    },
    {
      "epoch": 1.2906358224495802,
      "grad_norm": 0.9857951998710632,
      "learning_rate": 9.508777439336447e-06,
      "loss": 0.111,
      "num_input_tokens_seen": 45854712,
      "step": 28510
    },
    {
      "epoch": 1.290862174336223,
      "grad_norm": 0.7049841284751892,
      "learning_rate": 9.50107310123393e-06,
      "loss": 0.116,
      "num_input_tokens_seen": 45862520,
      "step": 28515
    },
    {
      "epoch": 1.291088526222866,
      "grad_norm": 0.659123957157135,
      "learning_rate": 9.493371153260702e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 45870424,
      "step": 28520
    },
    {
      "epoch": 1.291314878109509,
      "grad_norm": 0.42103976011276245,
      "learning_rate": 9.485671596604523e-06,
      "loss": 0.1373,
      "num_input_tokens_seen": 45877944,
      "step": 28525
    },
    {
      "epoch": 1.291541229996152,
      "grad_norm": 0.4764624536037445,
      "learning_rate": 9.477974432452738e-06,
      "loss": 0.1505,
      "num_input_tokens_seen": 45886296,
      "step": 28530
    },
    {
      "epoch": 1.291767581882795,
      "grad_norm": 0.3445793390274048,
      "learning_rate": 9.470279661992356e-06,
      "loss": 0.1231,
      "num_input_tokens_seen": 45894968,
      "step": 28535
    },
    {
      "epoch": 1.291993933769438,
      "grad_norm": 0.6266487836837769,
      "learning_rate": 9.462587286410021e-06,
      "loss": 0.144,
      "num_input_tokens_seen": 45903000,
      "step": 28540
    },
    {
      "epoch": 1.292220285656081,
      "grad_norm": 0.3654532730579376,
      "learning_rate": 9.454897306891972e-06,
      "loss": 0.144,
      "num_input_tokens_seen": 45910168,
      "step": 28545
    },
    {
      "epoch": 1.292446637542724,
      "grad_norm": 0.667283296585083,
      "learning_rate": 9.44720972462411e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 45918904,
      "step": 28550
    },
    {
      "epoch": 1.292672989429367,
      "grad_norm": 0.6129404902458191,
      "learning_rate": 9.439524540791964e-06,
      "loss": 0.1689,
      "num_input_tokens_seen": 45926680,
      "step": 28555
    },
    {
      "epoch": 1.29289934131601,
      "grad_norm": 0.8135872483253479,
      "learning_rate": 9.431841756580673e-06,
      "loss": 0.1332,
      "num_input_tokens_seen": 45934904,
      "step": 28560
    },
    {
      "epoch": 1.293125693202653,
      "grad_norm": 1.2040185928344727,
      "learning_rate": 9.42416137317503e-06,
      "loss": 0.1611,
      "num_input_tokens_seen": 45942552,
      "step": 28565
    },
    {
      "epoch": 1.293352045089296,
      "grad_norm": 0.3857024312019348,
      "learning_rate": 9.416483391759437e-06,
      "loss": 0.1536,
      "num_input_tokens_seen": 45950104,
      "step": 28570
    },
    {
      "epoch": 1.293578396975939,
      "grad_norm": 1.1245888471603394,
      "learning_rate": 9.408807813517945e-06,
      "loss": 0.1872,
      "num_input_tokens_seen": 45958104,
      "step": 28575
    },
    {
      "epoch": 1.293804748862582,
      "grad_norm": 0.5216441750526428,
      "learning_rate": 9.401134639634221e-06,
      "loss": 0.1344,
      "num_input_tokens_seen": 45966360,
      "step": 28580
    },
    {
      "epoch": 1.2940311007492247,
      "grad_norm": 0.8270658254623413,
      "learning_rate": 9.393463871291555e-06,
      "loss": 0.1447,
      "num_input_tokens_seen": 45974232,
      "step": 28585
    },
    {
      "epoch": 1.2942574526358677,
      "grad_norm": 1.2627332210540771,
      "learning_rate": 9.385795509672881e-06,
      "loss": 0.1264,
      "num_input_tokens_seen": 45982200,
      "step": 28590
    },
    {
      "epoch": 1.2944838045225107,
      "grad_norm": 0.8505100607872009,
      "learning_rate": 9.378129555960771e-06,
      "loss": 0.1974,
      "num_input_tokens_seen": 45989592,
      "step": 28595
    },
    {
      "epoch": 1.2947101564091537,
      "grad_norm": 0.5203337669372559,
      "learning_rate": 9.370466011337392e-06,
      "loss": 0.1368,
      "num_input_tokens_seen": 45998328,
      "step": 28600
    },
    {
      "epoch": 1.2947101564091537,
      "eval_loss": 0.1445377618074417,
      "eval_runtime": 405.6948,
      "eval_samples_per_second": 96.799,
      "eval_steps_per_second": 24.2,
      "num_input_tokens_seen": 45998328,
      "step": 28600
    },
    {
      "epoch": 1.2949365082957967,
      "grad_norm": 0.6904234290122986,
      "learning_rate": 9.362804876984573e-06,
      "loss": 0.1404,
      "num_input_tokens_seen": 46005976,
      "step": 28605
    },
    {
      "epoch": 1.2951628601824396,
      "grad_norm": 0.6887508630752563,
      "learning_rate": 9.355146154083747e-06,
      "loss": 0.1497,
      "num_input_tokens_seen": 46014136,
      "step": 28610
    },
    {
      "epoch": 1.2953892120690826,
      "grad_norm": 1.3906841278076172,
      "learning_rate": 9.347489843815987e-06,
      "loss": 0.1679,
      "num_input_tokens_seen": 46022936,
      "step": 28615
    },
    {
      "epoch": 1.2956155639557256,
      "grad_norm": 0.9160636067390442,
      "learning_rate": 9.339835947362002e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 46030616,
      "step": 28620
    },
    {
      "epoch": 1.2958419158423684,
      "grad_norm": 1.0104212760925293,
      "learning_rate": 9.332184465902105e-06,
      "loss": 0.1618,
      "num_input_tokens_seen": 46038392,
      "step": 28625
    },
    {
      "epoch": 1.2960682677290114,
      "grad_norm": 0.5734822750091553,
      "learning_rate": 9.324535400616266e-06,
      "loss": 0.1581,
      "num_input_tokens_seen": 46046328,
      "step": 28630
    },
    {
      "epoch": 1.2962946196156544,
      "grad_norm": 0.7393149733543396,
      "learning_rate": 9.31688875268405e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 46054200,
      "step": 28635
    },
    {
      "epoch": 1.2965209715022974,
      "grad_norm": 0.6353784203529358,
      "learning_rate": 9.309244523284674e-06,
      "loss": 0.1825,
      "num_input_tokens_seen": 46062360,
      "step": 28640
    },
    {
      "epoch": 1.2967473233889404,
      "grad_norm": 0.31883078813552856,
      "learning_rate": 9.301602713596982e-06,
      "loss": 0.1259,
      "num_input_tokens_seen": 46070264,
      "step": 28645
    },
    {
      "epoch": 1.2969736752755834,
      "grad_norm": 0.4258585274219513,
      "learning_rate": 9.293963324799432e-06,
      "loss": 0.1643,
      "num_input_tokens_seen": 46078808,
      "step": 28650
    },
    {
      "epoch": 1.2972000271622264,
      "grad_norm": 0.36919447779655457,
      "learning_rate": 9.286326358070104e-06,
      "loss": 0.089,
      "num_input_tokens_seen": 46086360,
      "step": 28655
    },
    {
      "epoch": 1.2974263790488694,
      "grad_norm": 0.4781053960323334,
      "learning_rate": 9.278691814586729e-06,
      "loss": 0.1223,
      "num_input_tokens_seen": 46094488,
      "step": 28660
    },
    {
      "epoch": 1.2976527309355124,
      "grad_norm": 0.7152078747749329,
      "learning_rate": 9.271059695526635e-06,
      "loss": 0.1795,
      "num_input_tokens_seen": 46102296,
      "step": 28665
    },
    {
      "epoch": 1.2978790828221554,
      "grad_norm": 0.46377602219581604,
      "learning_rate": 9.263430002066805e-06,
      "loss": 0.1442,
      "num_input_tokens_seen": 46110360,
      "step": 28670
    },
    {
      "epoch": 1.2981054347087984,
      "grad_norm": 0.5107092261314392,
      "learning_rate": 9.25580273538382e-06,
      "loss": 0.1483,
      "num_input_tokens_seen": 46117880,
      "step": 28675
    },
    {
      "epoch": 1.2983317865954414,
      "grad_norm": 0.551502525806427,
      "learning_rate": 9.248177896653907e-06,
      "loss": 0.1611,
      "num_input_tokens_seen": 46125176,
      "step": 28680
    },
    {
      "epoch": 1.2985581384820843,
      "grad_norm": 1.6884748935699463,
      "learning_rate": 9.240555487052918e-06,
      "loss": 0.1332,
      "num_input_tokens_seen": 46133016,
      "step": 28685
    },
    {
      "epoch": 1.2987844903687273,
      "grad_norm": 0.5943697690963745,
      "learning_rate": 9.232935507756313e-06,
      "loss": 0.152,
      "num_input_tokens_seen": 46140568,
      "step": 28690
    },
    {
      "epoch": 1.2990108422553703,
      "grad_norm": 0.8053395748138428,
      "learning_rate": 9.225317959939193e-06,
      "loss": 0.2096,
      "num_input_tokens_seen": 46149048,
      "step": 28695
    },
    {
      "epoch": 1.299237194142013,
      "grad_norm": 0.5707180500030518,
      "learning_rate": 9.217702844776287e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 46156504,
      "step": 28700
    },
    {
      "epoch": 1.299463546028656,
      "grad_norm": 0.9864704608917236,
      "learning_rate": 9.210090163441929e-06,
      "loss": 0.1415,
      "num_input_tokens_seen": 46164952,
      "step": 28705
    },
    {
      "epoch": 1.299689897915299,
      "grad_norm": 0.4505701959133148,
      "learning_rate": 9.202479917110105e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 46172856,
      "step": 28710
    },
    {
      "epoch": 1.299916249801942,
      "grad_norm": 0.9231533408164978,
      "learning_rate": 9.194872106954392e-06,
      "loss": 0.1285,
      "num_input_tokens_seen": 46181304,
      "step": 28715
    },
    {
      "epoch": 1.300142601688585,
      "grad_norm": 0.8810542821884155,
      "learning_rate": 9.187266734148029e-06,
      "loss": 0.1166,
      "num_input_tokens_seen": 46189048,
      "step": 28720
    },
    {
      "epoch": 1.300368953575228,
      "grad_norm": 0.7522144913673401,
      "learning_rate": 9.179663799863849e-06,
      "loss": 0.1468,
      "num_input_tokens_seen": 46196344,
      "step": 28725
    },
    {
      "epoch": 1.300595305461871,
      "grad_norm": 0.5065655708312988,
      "learning_rate": 9.172063305274317e-06,
      "loss": 0.1317,
      "num_input_tokens_seen": 46203960,
      "step": 28730
    },
    {
      "epoch": 1.300821657348514,
      "grad_norm": 0.6782070398330688,
      "learning_rate": 9.164465251551527e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 46211832,
      "step": 28735
    },
    {
      "epoch": 1.3010480092351568,
      "grad_norm": 0.5341418385505676,
      "learning_rate": 9.156869639867205e-06,
      "loss": 0.1575,
      "num_input_tokens_seen": 46220056,
      "step": 28740
    },
    {
      "epoch": 1.3012743611217998,
      "grad_norm": 0.5292332172393799,
      "learning_rate": 9.149276471392677e-06,
      "loss": 0.1576,
      "num_input_tokens_seen": 46228216,
      "step": 28745
    },
    {
      "epoch": 1.3015007130084428,
      "grad_norm": 0.4635598659515381,
      "learning_rate": 9.141685747298914e-06,
      "loss": 0.1165,
      "num_input_tokens_seen": 46236088,
      "step": 28750
    },
    {
      "epoch": 1.3017270648950858,
      "grad_norm": 0.42785874009132385,
      "learning_rate": 9.13409746875649e-06,
      "loss": 0.1393,
      "num_input_tokens_seen": 46243768,
      "step": 28755
    },
    {
      "epoch": 1.3019534167817288,
      "grad_norm": 0.41184768080711365,
      "learning_rate": 9.12651163693562e-06,
      "loss": 0.1612,
      "num_input_tokens_seen": 46251288,
      "step": 28760
    },
    {
      "epoch": 1.3021797686683718,
      "grad_norm": 1.1024168729782104,
      "learning_rate": 9.11892825300614e-06,
      "loss": 0.1558,
      "num_input_tokens_seen": 46258872,
      "step": 28765
    },
    {
      "epoch": 1.3024061205550148,
      "grad_norm": 0.4336613416671753,
      "learning_rate": 9.111347318137491e-06,
      "loss": 0.1258,
      "num_input_tokens_seen": 46266840,
      "step": 28770
    },
    {
      "epoch": 1.3026324724416578,
      "grad_norm": 0.8059686422348022,
      "learning_rate": 9.103768833498755e-06,
      "loss": 0.0982,
      "num_input_tokens_seen": 46274648,
      "step": 28775
    },
    {
      "epoch": 1.3028588243283008,
      "grad_norm": 0.700011670589447,
      "learning_rate": 9.096192800258639e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 46283160,
      "step": 28780
    },
    {
      "epoch": 1.3030851762149438,
      "grad_norm": 0.4482572078704834,
      "learning_rate": 9.088619219585443e-06,
      "loss": 0.1476,
      "num_input_tokens_seen": 46291384,
      "step": 28785
    },
    {
      "epoch": 1.3033115281015868,
      "grad_norm": 0.49526140093803406,
      "learning_rate": 9.081048092647127e-06,
      "loss": 0.1346,
      "num_input_tokens_seen": 46298936,
      "step": 28790
    },
    {
      "epoch": 1.3035378799882298,
      "grad_norm": 0.6485099196434021,
      "learning_rate": 9.073479420611245e-06,
      "loss": 0.1352,
      "num_input_tokens_seen": 46306520,
      "step": 28795
    },
    {
      "epoch": 1.3037642318748728,
      "grad_norm": 0.34731951355934143,
      "learning_rate": 9.065913204644974e-06,
      "loss": 0.193,
      "num_input_tokens_seen": 46314552,
      "step": 28800
    },
    {
      "epoch": 1.3037642318748728,
      "eval_loss": 0.14402663707733154,
      "eval_runtime": 405.4082,
      "eval_samples_per_second": 96.868,
      "eval_steps_per_second": 24.218,
      "num_input_tokens_seen": 46314552,
      "step": 28800
    },
    {
      "epoch": 1.3039905837615158,
      "grad_norm": 0.5260813236236572,
      "learning_rate": 9.058349445915135e-06,
      "loss": 0.1333,
      "num_input_tokens_seen": 46322616,
      "step": 28805
    },
    {
      "epoch": 1.3042169356481585,
      "grad_norm": 0.6019811630249023,
      "learning_rate": 9.050788145588138e-06,
      "loss": 0.1663,
      "num_input_tokens_seen": 46330968,
      "step": 28810
    },
    {
      "epoch": 1.3044432875348015,
      "grad_norm": 0.9367200136184692,
      "learning_rate": 9.043229304830039e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 46339704,
      "step": 28815
    },
    {
      "epoch": 1.3046696394214445,
      "grad_norm": 0.6925216913223267,
      "learning_rate": 9.035672924806515e-06,
      "loss": 0.114,
      "num_input_tokens_seen": 46349080,
      "step": 28820
    },
    {
      "epoch": 1.3048959913080875,
      "grad_norm": 0.8792828917503357,
      "learning_rate": 9.028119006682839e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 46357048,
      "step": 28825
    },
    {
      "epoch": 1.3051223431947305,
      "grad_norm": 1.0145487785339355,
      "learning_rate": 9.020567551623935e-06,
      "loss": 0.1485,
      "num_input_tokens_seen": 46364760,
      "step": 28830
    },
    {
      "epoch": 1.3053486950813735,
      "grad_norm": 0.4159623980522156,
      "learning_rate": 9.013018560794318e-06,
      "loss": 0.1191,
      "num_input_tokens_seen": 46372984,
      "step": 28835
    },
    {
      "epoch": 1.3055750469680165,
      "grad_norm": 0.804953396320343,
      "learning_rate": 9.005472035358139e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 46380664,
      "step": 28840
    },
    {
      "epoch": 1.3058013988546595,
      "grad_norm": 0.42174628376960754,
      "learning_rate": 8.997927976479185e-06,
      "loss": 0.1276,
      "num_input_tokens_seen": 46389176,
      "step": 28845
    },
    {
      "epoch": 1.3060277507413025,
      "grad_norm": 0.9142602682113647,
      "learning_rate": 8.99038638532082e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 46398168,
      "step": 28850
    },
    {
      "epoch": 1.3062541026279453,
      "grad_norm": 1.187580943107605,
      "learning_rate": 8.982847263046065e-06,
      "loss": 0.1651,
      "num_input_tokens_seen": 46405560,
      "step": 28855
    },
    {
      "epoch": 1.3064804545145883,
      "grad_norm": 0.49636274576187134,
      "learning_rate": 8.975310610817555e-06,
      "loss": 0.1268,
      "num_input_tokens_seen": 46413432,
      "step": 28860
    },
    {
      "epoch": 1.3067068064012313,
      "grad_norm": 1.2448533773422241,
      "learning_rate": 8.967776429797528e-06,
      "loss": 0.131,
      "num_input_tokens_seen": 46421976,
      "step": 28865
    },
    {
      "epoch": 1.3069331582878743,
      "grad_norm": 0.52447509765625,
      "learning_rate": 8.960244721147842e-06,
      "loss": 0.1447,
      "num_input_tokens_seen": 46430008,
      "step": 28870
    },
    {
      "epoch": 1.3071595101745173,
      "grad_norm": 0.5531304478645325,
      "learning_rate": 8.952715486029995e-06,
      "loss": 0.1931,
      "num_input_tokens_seen": 46438328,
      "step": 28875
    },
    {
      "epoch": 1.3073858620611603,
      "grad_norm": 1.261103630065918,
      "learning_rate": 8.945188725605075e-06,
      "loss": 0.2022,
      "num_input_tokens_seen": 46446008,
      "step": 28880
    },
    {
      "epoch": 1.3076122139478032,
      "grad_norm": 0.4565802216529846,
      "learning_rate": 8.937664441033817e-06,
      "loss": 0.104,
      "num_input_tokens_seen": 46454584,
      "step": 28885
    },
    {
      "epoch": 1.3078385658344462,
      "grad_norm": 0.893592894077301,
      "learning_rate": 8.930142633476549e-06,
      "loss": 0.1292,
      "num_input_tokens_seen": 46462584,
      "step": 28890
    },
    {
      "epoch": 1.3080649177210892,
      "grad_norm": 0.43328142166137695,
      "learning_rate": 8.92262330409323e-06,
      "loss": 0.1471,
      "num_input_tokens_seen": 46471224,
      "step": 28895
    },
    {
      "epoch": 1.3082912696077322,
      "grad_norm": 0.7462189793586731,
      "learning_rate": 8.915106454043448e-06,
      "loss": 0.162,
      "num_input_tokens_seen": 46478968,
      "step": 28900
    },
    {
      "epoch": 1.3085176214943752,
      "grad_norm": 0.44212836027145386,
      "learning_rate": 8.90759208448638e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 46486648,
      "step": 28905
    },
    {
      "epoch": 1.3087439733810182,
      "grad_norm": 0.3226107954978943,
      "learning_rate": 8.900080196580848e-06,
      "loss": 0.1322,
      "num_input_tokens_seen": 46495768,
      "step": 28910
    },
    {
      "epoch": 1.3089703252676612,
      "grad_norm": 0.5212305784225464,
      "learning_rate": 8.892570791485267e-06,
      "loss": 0.116,
      "num_input_tokens_seen": 46503736,
      "step": 28915
    },
    {
      "epoch": 1.3091966771543042,
      "grad_norm": 0.6609404683113098,
      "learning_rate": 8.885063870357688e-06,
      "loss": 0.1412,
      "num_input_tokens_seen": 46511640,
      "step": 28920
    },
    {
      "epoch": 1.309423029040947,
      "grad_norm": 0.6737493872642517,
      "learning_rate": 8.87755943435578e-06,
      "loss": 0.1313,
      "num_input_tokens_seen": 46519896,
      "step": 28925
    },
    {
      "epoch": 1.30964938092759,
      "grad_norm": 0.32332247495651245,
      "learning_rate": 8.87005748463681e-06,
      "loss": 0.1064,
      "num_input_tokens_seen": 46527832,
      "step": 28930
    },
    {
      "epoch": 1.309875732814233,
      "grad_norm": 0.627981960773468,
      "learning_rate": 8.862558022357681e-06,
      "loss": 0.1662,
      "num_input_tokens_seen": 46536568,
      "step": 28935
    },
    {
      "epoch": 1.310102084700876,
      "grad_norm": 1.1130313873291016,
      "learning_rate": 8.855061048674903e-06,
      "loss": 0.1445,
      "num_input_tokens_seen": 46544184,
      "step": 28940
    },
    {
      "epoch": 1.310328436587519,
      "grad_norm": 0.6420381665229797,
      "learning_rate": 8.847566564744595e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 46552440,
      "step": 28945
    },
    {
      "epoch": 1.310554788474162,
      "grad_norm": 0.42361021041870117,
      "learning_rate": 8.840074571722512e-06,
      "loss": 0.1668,
      "num_input_tokens_seen": 46560120,
      "step": 28950
    },
    {
      "epoch": 1.310781140360805,
      "grad_norm": 0.5076321363449097,
      "learning_rate": 8.832585070764002e-06,
      "loss": 0.1209,
      "num_input_tokens_seen": 46567544,
      "step": 28955
    },
    {
      "epoch": 1.311007492247448,
      "grad_norm": 0.8410394787788391,
      "learning_rate": 8.825098063024045e-06,
      "loss": 0.1472,
      "num_input_tokens_seen": 46576088,
      "step": 28960
    },
    {
      "epoch": 1.3112338441340907,
      "grad_norm": 0.553973913192749,
      "learning_rate": 8.817613549657244e-06,
      "loss": 0.1467,
      "num_input_tokens_seen": 46584408,
      "step": 28965
    },
    {
      "epoch": 1.3114601960207337,
      "grad_norm": 0.3911024332046509,
      "learning_rate": 8.810131531817783e-06,
      "loss": 0.0989,
      "num_input_tokens_seen": 46592632,
      "step": 28970
    },
    {
      "epoch": 1.3116865479073767,
      "grad_norm": 0.4296053647994995,
      "learning_rate": 8.802652010659496e-06,
      "loss": 0.1215,
      "num_input_tokens_seen": 46600856,
      "step": 28975
    },
    {
      "epoch": 1.3119128997940197,
      "grad_norm": 0.5995635390281677,
      "learning_rate": 8.795174987335827e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 46608152,
      "step": 28980
    },
    {
      "epoch": 1.3121392516806627,
      "grad_norm": 0.846699059009552,
      "learning_rate": 8.787700462999807e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 46615960,
      "step": 28985
    },
    {
      "epoch": 1.3123656035673057,
      "grad_norm": 0.6377202868461609,
      "learning_rate": 8.780228438804122e-06,
      "loss": 0.1676,
      "num_input_tokens_seen": 46623672,
      "step": 28990
    },
    {
      "epoch": 1.3125919554539487,
      "grad_norm": 0.4974634647369385,
      "learning_rate": 8.772758915901032e-06,
      "loss": 0.1133,
      "num_input_tokens_seen": 46631640,
      "step": 28995
    },
    {
      "epoch": 1.3128183073405917,
      "grad_norm": 0.7724976539611816,
      "learning_rate": 8.765291895442443e-06,
      "loss": 0.1264,
      "num_input_tokens_seen": 46639640,
      "step": 29000
    },
    {
      "epoch": 1.3128183073405917,
      "eval_loss": 0.14409764111042023,
      "eval_runtime": 405.0565,
      "eval_samples_per_second": 96.952,
      "eval_steps_per_second": 24.239,
      "num_input_tokens_seen": 46639640,
      "step": 29000
    },
    {
      "epoch": 1.3130446592272347,
      "grad_norm": 0.23791556060314178,
      "learning_rate": 8.75782737857987e-06,
      "loss": 0.1163,
      "num_input_tokens_seen": 46647480,
      "step": 29005
    },
    {
      "epoch": 1.3132710111138777,
      "grad_norm": 0.5392873287200928,
      "learning_rate": 8.750365366464425e-06,
      "loss": 0.166,
      "num_input_tokens_seen": 46654904,
      "step": 29010
    },
    {
      "epoch": 1.3134973630005207,
      "grad_norm": 0.429718017578125,
      "learning_rate": 8.742905860246838e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 46663832,
      "step": 29015
    },
    {
      "epoch": 1.3137237148871637,
      "grad_norm": 0.46087735891342163,
      "learning_rate": 8.735448861077478e-06,
      "loss": 0.1343,
      "num_input_tokens_seen": 46672696,
      "step": 29020
    },
    {
      "epoch": 1.3139500667738067,
      "grad_norm": 0.5610571503639221,
      "learning_rate": 8.727994370106288e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 46680920,
      "step": 29025
    },
    {
      "epoch": 1.3141764186604497,
      "grad_norm": 0.655487596988678,
      "learning_rate": 8.720542388482861e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 46688760,
      "step": 29030
    },
    {
      "epoch": 1.3144027705470926,
      "grad_norm": 0.6675741076469421,
      "learning_rate": 8.71309291735637e-06,
      "loss": 0.1555,
      "num_input_tokens_seen": 46696280,
      "step": 29035
    },
    {
      "epoch": 1.3146291224337354,
      "grad_norm": 0.4205317497253418,
      "learning_rate": 8.705645957875621e-06,
      "loss": 0.1473,
      "num_input_tokens_seen": 46704312,
      "step": 29040
    },
    {
      "epoch": 1.3148554743203784,
      "grad_norm": 0.5835952162742615,
      "learning_rate": 8.698201511189048e-06,
      "loss": 0.15,
      "num_input_tokens_seen": 46713688,
      "step": 29045
    },
    {
      "epoch": 1.3150818262070214,
      "grad_norm": 0.49791428446769714,
      "learning_rate": 8.690759578444649e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 46721784,
      "step": 29050
    },
    {
      "epoch": 1.3153081780936644,
      "grad_norm": 0.9293984174728394,
      "learning_rate": 8.68332016079008e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 46729496,
      "step": 29055
    },
    {
      "epoch": 1.3155345299803074,
      "grad_norm": 0.3969249725341797,
      "learning_rate": 8.6758832593726e-06,
      "loss": 0.1461,
      "num_input_tokens_seen": 46737464,
      "step": 29060
    },
    {
      "epoch": 1.3157608818669504,
      "grad_norm": 0.8355594873428345,
      "learning_rate": 8.668448875339053e-06,
      "loss": 0.1503,
      "num_input_tokens_seen": 46744824,
      "step": 29065
    },
    {
      "epoch": 1.3159872337535934,
      "grad_norm": 1.1107908487319946,
      "learning_rate": 8.661017009835933e-06,
      "loss": 0.147,
      "num_input_tokens_seen": 46752920,
      "step": 29070
    },
    {
      "epoch": 1.3162135856402364,
      "grad_norm": 0.6192798614501953,
      "learning_rate": 8.653587664009311e-06,
      "loss": 0.1393,
      "num_input_tokens_seen": 46760568,
      "step": 29075
    },
    {
      "epoch": 1.3164399375268792,
      "grad_norm": 0.3821396827697754,
      "learning_rate": 8.646160839004902e-06,
      "loss": 0.1778,
      "num_input_tokens_seen": 46768568,
      "step": 29080
    },
    {
      "epoch": 1.3166662894135222,
      "grad_norm": 1.1655287742614746,
      "learning_rate": 8.638736535967998e-06,
      "loss": 0.1421,
      "num_input_tokens_seen": 46777016,
      "step": 29085
    },
    {
      "epoch": 1.3168926413001651,
      "grad_norm": 1.0934648513793945,
      "learning_rate": 8.631314756043535e-06,
      "loss": 0.1762,
      "num_input_tokens_seen": 46784760,
      "step": 29090
    },
    {
      "epoch": 1.3171189931868081,
      "grad_norm": 0.44545015692710876,
      "learning_rate": 8.62389550037603e-06,
      "loss": 0.1676,
      "num_input_tokens_seen": 46793816,
      "step": 29095
    },
    {
      "epoch": 1.3173453450734511,
      "grad_norm": 0.6732417345046997,
      "learning_rate": 8.616478770109646e-06,
      "loss": 0.1532,
      "num_input_tokens_seen": 46801560,
      "step": 29100
    },
    {
      "epoch": 1.3175716969600941,
      "grad_norm": 0.6975809335708618,
      "learning_rate": 8.609064566388111e-06,
      "loss": 0.1286,
      "num_input_tokens_seen": 46809368,
      "step": 29105
    },
    {
      "epoch": 1.3177980488467371,
      "grad_norm": 0.4698946475982666,
      "learning_rate": 8.601652890354815e-06,
      "loss": 0.1422,
      "num_input_tokens_seen": 46817624,
      "step": 29110
    },
    {
      "epoch": 1.3180244007333801,
      "grad_norm": 1.5637104511260986,
      "learning_rate": 8.594243743152705e-06,
      "loss": 0.1624,
      "num_input_tokens_seen": 46826168,
      "step": 29115
    },
    {
      "epoch": 1.3182507526200231,
      "grad_norm": 0.45258840918540955,
      "learning_rate": 8.58683712592438e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 46833848,
      "step": 29120
    },
    {
      "epoch": 1.318477104506666,
      "grad_norm": 0.6669064164161682,
      "learning_rate": 8.579433039812037e-06,
      "loss": 0.1554,
      "num_input_tokens_seen": 46842040,
      "step": 29125
    },
    {
      "epoch": 1.318703456393309,
      "grad_norm": 0.9276596307754517,
      "learning_rate": 8.572031485957466e-06,
      "loss": 0.1624,
      "num_input_tokens_seen": 46850360,
      "step": 29130
    },
    {
      "epoch": 1.318929808279952,
      "grad_norm": 0.5230214595794678,
      "learning_rate": 8.564632465502084e-06,
      "loss": 0.1198,
      "num_input_tokens_seen": 46858712,
      "step": 29135
    },
    {
      "epoch": 1.319156160166595,
      "grad_norm": 0.5775060057640076,
      "learning_rate": 8.557235979586928e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 46867064,
      "step": 29140
    },
    {
      "epoch": 1.319382512053238,
      "grad_norm": 0.8736814260482788,
      "learning_rate": 8.549842029352606e-06,
      "loss": 0.1373,
      "num_input_tokens_seen": 46875128,
      "step": 29145
    },
    {
      "epoch": 1.3196088639398809,
      "grad_norm": 0.8085704445838928,
      "learning_rate": 8.542450615939376e-06,
      "loss": 0.1118,
      "num_input_tokens_seen": 46883096,
      "step": 29150
    },
    {
      "epoch": 1.3198352158265239,
      "grad_norm": 0.8048359751701355,
      "learning_rate": 8.535061740487082e-06,
      "loss": 0.1749,
      "num_input_tokens_seen": 46890904,
      "step": 29155
    },
    {
      "epoch": 1.3200615677131669,
      "grad_norm": 0.7596100568771362,
      "learning_rate": 8.527675404135168e-06,
      "loss": 0.1701,
      "num_input_tokens_seen": 46898712,
      "step": 29160
    },
    {
      "epoch": 1.3202879195998098,
      "grad_norm": 1.0446481704711914,
      "learning_rate": 8.520291608022724e-06,
      "loss": 0.138,
      "num_input_tokens_seen": 46906648,
      "step": 29165
    },
    {
      "epoch": 1.3205142714864528,
      "grad_norm": 0.7069670557975769,
      "learning_rate": 8.512910353288398e-06,
      "loss": 0.1357,
      "num_input_tokens_seen": 46914872,
      "step": 29170
    },
    {
      "epoch": 1.3207406233730958,
      "grad_norm": 0.47180742025375366,
      "learning_rate": 8.505531641070486e-06,
      "loss": 0.1411,
      "num_input_tokens_seen": 46922712,
      "step": 29175
    },
    {
      "epoch": 1.3209669752597388,
      "grad_norm": 0.8559341430664062,
      "learning_rate": 8.498155472506885e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 46930936,
      "step": 29180
    },
    {
      "epoch": 1.3211933271463818,
      "grad_norm": 0.4121089279651642,
      "learning_rate": 8.49078184873508e-06,
      "loss": 0.1184,
      "num_input_tokens_seen": 46939000,
      "step": 29185
    },
    {
      "epoch": 1.3214196790330248,
      "grad_norm": 0.3728634715080261,
      "learning_rate": 8.483410770892188e-06,
      "loss": 0.1428,
      "num_input_tokens_seen": 46946808,
      "step": 29190
    },
    {
      "epoch": 1.3216460309196676,
      "grad_norm": 0.8361465930938721,
      "learning_rate": 8.476042240114909e-06,
      "loss": 0.154,
      "num_input_tokens_seen": 46954488,
      "step": 29195
    },
    {
      "epoch": 1.3218723828063106,
      "grad_norm": 1.3334249258041382,
      "learning_rate": 8.468676257539568e-06,
      "loss": 0.1665,
      "num_input_tokens_seen": 46962616,
      "step": 29200
    },
    {
      "epoch": 1.3218723828063106,
      "eval_loss": 0.1440253108739853,
      "eval_runtime": 405.0887,
      "eval_samples_per_second": 96.944,
      "eval_steps_per_second": 24.237,
      "num_input_tokens_seen": 46962616,
      "step": 29200
    },
    {
      "epoch": 1.3220987346929536,
      "grad_norm": 0.4902951419353485,
      "learning_rate": 8.4613128243021e-06,
      "loss": 0.152,
      "num_input_tokens_seen": 46970232,
      "step": 29205
    },
    {
      "epoch": 1.3223250865795966,
      "grad_norm": 0.5254189968109131,
      "learning_rate": 8.453951941538028e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 46978552,
      "step": 29210
    },
    {
      "epoch": 1.3225514384662396,
      "grad_norm": 1.4380143880844116,
      "learning_rate": 8.446593610382495e-06,
      "loss": 0.141,
      "num_input_tokens_seen": 46986392,
      "step": 29215
    },
    {
      "epoch": 1.3227777903528826,
      "grad_norm": 1.5259356498718262,
      "learning_rate": 8.439237831970259e-06,
      "loss": 0.1786,
      "num_input_tokens_seen": 46994168,
      "step": 29220
    },
    {
      "epoch": 1.3230041422395256,
      "grad_norm": 0.38344478607177734,
      "learning_rate": 8.431884607435667e-06,
      "loss": 0.1349,
      "num_input_tokens_seen": 47002040,
      "step": 29225
    },
    {
      "epoch": 1.3232304941261686,
      "grad_norm": 1.3513752222061157,
      "learning_rate": 8.424533937912665e-06,
      "loss": 0.166,
      "num_input_tokens_seen": 47010232,
      "step": 29230
    },
    {
      "epoch": 1.3234568460128115,
      "grad_norm": 1.1098331212997437,
      "learning_rate": 8.41718582453484e-06,
      "loss": 0.1322,
      "num_input_tokens_seen": 47018232,
      "step": 29235
    },
    {
      "epoch": 1.3236831978994545,
      "grad_norm": 1.2950983047485352,
      "learning_rate": 8.409840268435346e-06,
      "loss": 0.1515,
      "num_input_tokens_seen": 47026168,
      "step": 29240
    },
    {
      "epoch": 1.3239095497860975,
      "grad_norm": 0.7332629561424255,
      "learning_rate": 8.402497270746976e-06,
      "loss": 0.1357,
      "num_input_tokens_seen": 47034200,
      "step": 29245
    },
    {
      "epoch": 1.3241359016727405,
      "grad_norm": 0.3475784957408905,
      "learning_rate": 8.395156832602095e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 47042520,
      "step": 29250
    },
    {
      "epoch": 1.3243622535593835,
      "grad_norm": 0.572532057762146,
      "learning_rate": 8.387818955132707e-06,
      "loss": 0.1319,
      "num_input_tokens_seen": 47050680,
      "step": 29255
    },
    {
      "epoch": 1.3245886054460265,
      "grad_norm": 0.89332515001297,
      "learning_rate": 8.38048363947039e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 47058648,
      "step": 29260
    },
    {
      "epoch": 1.3248149573326693,
      "grad_norm": 0.7814176678657532,
      "learning_rate": 8.373150886746351e-06,
      "loss": 0.1539,
      "num_input_tokens_seen": 47066872,
      "step": 29265
    },
    {
      "epoch": 1.3250413092193123,
      "grad_norm": 0.493221640586853,
      "learning_rate": 8.365820698091397e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 47074680,
      "step": 29270
    },
    {
      "epoch": 1.3252676611059553,
      "grad_norm": 0.6393760442733765,
      "learning_rate": 8.358493074635922e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 47082136,
      "step": 29275
    },
    {
      "epoch": 1.3254940129925983,
      "grad_norm": 1.1939358711242676,
      "learning_rate": 8.351168017509948e-06,
      "loss": 0.1482,
      "num_input_tokens_seen": 47091032,
      "step": 29280
    },
    {
      "epoch": 1.3257203648792413,
      "grad_norm": 0.6273094415664673,
      "learning_rate": 8.343845527843094e-06,
      "loss": 0.1295,
      "num_input_tokens_seen": 47099032,
      "step": 29285
    },
    {
      "epoch": 1.3259467167658843,
      "grad_norm": 0.2602040469646454,
      "learning_rate": 8.336525606764566e-06,
      "loss": 0.1137,
      "num_input_tokens_seen": 47107416,
      "step": 29290
    },
    {
      "epoch": 1.3261730686525273,
      "grad_norm": 0.6669437289237976,
      "learning_rate": 8.329208255403204e-06,
      "loss": 0.1925,
      "num_input_tokens_seen": 47116344,
      "step": 29295
    },
    {
      "epoch": 1.3263994205391703,
      "grad_norm": 0.5864677429199219,
      "learning_rate": 8.321893474887426e-06,
      "loss": 0.1924,
      "num_input_tokens_seen": 47124376,
      "step": 29300
    },
    {
      "epoch": 1.326625772425813,
      "grad_norm": 0.4322768747806549,
      "learning_rate": 8.31458126634526e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 47132376,
      "step": 29305
    },
    {
      "epoch": 1.326852124312456,
      "grad_norm": 0.689971387386322,
      "learning_rate": 8.30727163090435e-06,
      "loss": 0.0824,
      "num_input_tokens_seen": 47140280,
      "step": 29310
    },
    {
      "epoch": 1.327078476199099,
      "grad_norm": 0.8483372330665588,
      "learning_rate": 8.29996456969192e-06,
      "loss": 0.1165,
      "num_input_tokens_seen": 47148120,
      "step": 29315
    },
    {
      "epoch": 1.327304828085742,
      "grad_norm": 0.3869923949241638,
      "learning_rate": 8.292660083834818e-06,
      "loss": 0.1131,
      "num_input_tokens_seen": 47155736,
      "step": 29320
    },
    {
      "epoch": 1.327531179972385,
      "grad_norm": 0.6701510548591614,
      "learning_rate": 8.2853581744595e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 47163736,
      "step": 29325
    },
    {
      "epoch": 1.327757531859028,
      "grad_norm": 0.6405420899391174,
      "learning_rate": 8.278058842691991e-06,
      "loss": 0.15,
      "num_input_tokens_seen": 47171608,
      "step": 29330
    },
    {
      "epoch": 1.327983883745671,
      "grad_norm": 0.6370717883110046,
      "learning_rate": 8.27076208965796e-06,
      "loss": 0.1173,
      "num_input_tokens_seen": 47179448,
      "step": 29335
    },
    {
      "epoch": 1.328210235632314,
      "grad_norm": 0.8057785630226135,
      "learning_rate": 8.263467916482637e-06,
      "loss": 0.1317,
      "num_input_tokens_seen": 47187192,
      "step": 29340
    },
    {
      "epoch": 1.328436587518957,
      "grad_norm": 0.48884862661361694,
      "learning_rate": 8.256176324290885e-06,
      "loss": 0.1545,
      "num_input_tokens_seen": 47195480,
      "step": 29345
    },
    {
      "epoch": 1.3286629394056,
      "grad_norm": 0.5097073912620544,
      "learning_rate": 8.248887314207168e-06,
      "loss": 0.1625,
      "num_input_tokens_seen": 47203128,
      "step": 29350
    },
    {
      "epoch": 1.328889291292243,
      "grad_norm": 0.7388318777084351,
      "learning_rate": 8.24160088735553e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 47212216,
      "step": 29355
    },
    {
      "epoch": 1.329115643178886,
      "grad_norm": 0.5784687995910645,
      "learning_rate": 8.234317044859629e-06,
      "loss": 0.1314,
      "num_input_tokens_seen": 47220056,
      "step": 29360
    },
    {
      "epoch": 1.329341995065529,
      "grad_norm": 0.5210086107254028,
      "learning_rate": 8.227035787842744e-06,
      "loss": 0.1552,
      "num_input_tokens_seen": 47227736,
      "step": 29365
    },
    {
      "epoch": 1.329568346952172,
      "grad_norm": 1.395948886871338,
      "learning_rate": 8.219757117427721e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 47235672,
      "step": 29370
    },
    {
      "epoch": 1.329794698838815,
      "grad_norm": 0.6872913837432861,
      "learning_rate": 8.212481034737014e-06,
      "loss": 0.1211,
      "num_input_tokens_seen": 47243768,
      "step": 29375
    },
    {
      "epoch": 1.3300210507254577,
      "grad_norm": 0.9154176115989685,
      "learning_rate": 8.205207540892707e-06,
      "loss": 0.1646,
      "num_input_tokens_seen": 47251960,
      "step": 29380
    },
    {
      "epoch": 1.3302474026121007,
      "grad_norm": 0.5048527717590332,
      "learning_rate": 8.197936637016442e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 47260376,
      "step": 29385
    },
    {
      "epoch": 1.3304737544987437,
      "grad_norm": 1.3476085662841797,
      "learning_rate": 8.190668324229508e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 47268184,
      "step": 29390
    },
    {
      "epoch": 1.3307001063853867,
      "grad_norm": 0.6013070940971375,
      "learning_rate": 8.183402603652749e-06,
      "loss": 0.1394,
      "num_input_tokens_seen": 47276280,
      "step": 29395
    },
    {
      "epoch": 1.3309264582720297,
      "grad_norm": 0.7323324084281921,
      "learning_rate": 8.176139476406635e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 47284152,
      "step": 29400
    },
    {
      "epoch": 1.3309264582720297,
      "eval_loss": 0.14406141638755798,
      "eval_runtime": 404.8991,
      "eval_samples_per_second": 96.99,
      "eval_steps_per_second": 24.248,
      "num_input_tokens_seen": 47284152,
      "step": 29400
    },
    {
      "epoch": 1.3311528101586727,
      "grad_norm": 0.6029396653175354,
      "learning_rate": 8.16887894361125e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 47291576,
      "step": 29405
    },
    {
      "epoch": 1.3313791620453157,
      "grad_norm": 0.8879518508911133,
      "learning_rate": 8.161621006386233e-06,
      "loss": 0.1617,
      "num_input_tokens_seen": 47299000,
      "step": 29410
    },
    {
      "epoch": 1.3316055139319587,
      "grad_norm": 0.3496737778186798,
      "learning_rate": 8.154365665850869e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 47307416,
      "step": 29415
    },
    {
      "epoch": 1.3318318658186015,
      "grad_norm": 0.24375994503498077,
      "learning_rate": 8.147112923124005e-06,
      "loss": 0.1094,
      "num_input_tokens_seen": 47314904,
      "step": 29420
    },
    {
      "epoch": 1.3320582177052445,
      "grad_norm": 1.167655110359192,
      "learning_rate": 8.13986277932412e-06,
      "loss": 0.1208,
      "num_input_tokens_seen": 47322552,
      "step": 29425
    },
    {
      "epoch": 1.3322845695918875,
      "grad_norm": 0.6975271105766296,
      "learning_rate": 8.132615235569277e-06,
      "loss": 0.1261,
      "num_input_tokens_seen": 47330712,
      "step": 29430
    },
    {
      "epoch": 1.3325109214785305,
      "grad_norm": 1.0002018213272095,
      "learning_rate": 8.125370292977124e-06,
      "loss": 0.1199,
      "num_input_tokens_seen": 47338648,
      "step": 29435
    },
    {
      "epoch": 1.3327372733651734,
      "grad_norm": 1.0759817361831665,
      "learning_rate": 8.118127952664944e-06,
      "loss": 0.1764,
      "num_input_tokens_seen": 47346680,
      "step": 29440
    },
    {
      "epoch": 1.3329636252518164,
      "grad_norm": 0.3807688355445862,
      "learning_rate": 8.110888215749574e-06,
      "loss": 0.1316,
      "num_input_tokens_seen": 47353976,
      "step": 29445
    },
    {
      "epoch": 1.3331899771384594,
      "grad_norm": 0.44402384757995605,
      "learning_rate": 8.10365108334749e-06,
      "loss": 0.1442,
      "num_input_tokens_seen": 47363128,
      "step": 29450
    },
    {
      "epoch": 1.3334163290251024,
      "grad_norm": 0.4489436447620392,
      "learning_rate": 8.096416556574743e-06,
      "loss": 0.1111,
      "num_input_tokens_seen": 47370488,
      "step": 29455
    },
    {
      "epoch": 1.3336426809117454,
      "grad_norm": 1.0948299169540405,
      "learning_rate": 8.08918463654698e-06,
      "loss": 0.171,
      "num_input_tokens_seen": 47378072,
      "step": 29460
    },
    {
      "epoch": 1.3338690327983884,
      "grad_norm": 0.9040001630783081,
      "learning_rate": 8.081955324379458e-06,
      "loss": 0.1066,
      "num_input_tokens_seen": 47386616,
      "step": 29465
    },
    {
      "epoch": 1.3340953846850314,
      "grad_norm": 0.5674296617507935,
      "learning_rate": 8.074728621187039e-06,
      "loss": 0.1527,
      "num_input_tokens_seen": 47394520,
      "step": 29470
    },
    {
      "epoch": 1.3343217365716744,
      "grad_norm": 0.47541674971580505,
      "learning_rate": 8.067504528084158e-06,
      "loss": 0.1084,
      "num_input_tokens_seen": 47402936,
      "step": 29475
    },
    {
      "epoch": 1.3345480884583174,
      "grad_norm": 0.49777039885520935,
      "learning_rate": 8.060283046184861e-06,
      "loss": 0.1844,
      "num_input_tokens_seen": 47410360,
      "step": 29480
    },
    {
      "epoch": 1.3347744403449604,
      "grad_norm": 1.1009321212768555,
      "learning_rate": 8.053064176602806e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 47417912,
      "step": 29485
    },
    {
      "epoch": 1.3350007922316032,
      "grad_norm": 1.254056692123413,
      "learning_rate": 8.045847920451216e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 47425624,
      "step": 29490
    },
    {
      "epoch": 1.3352271441182462,
      "grad_norm": 0.5781404972076416,
      "learning_rate": 8.038634278842944e-06,
      "loss": 0.1748,
      "num_input_tokens_seen": 47433816,
      "step": 29495
    },
    {
      "epoch": 1.3354534960048892,
      "grad_norm": 0.8198229670524597,
      "learning_rate": 8.031423252890408e-06,
      "loss": 0.1622,
      "num_input_tokens_seen": 47442488,
      "step": 29500
    },
    {
      "epoch": 1.3356798478915322,
      "grad_norm": 0.4348616302013397,
      "learning_rate": 8.024214843705646e-06,
      "loss": 0.1722,
      "num_input_tokens_seen": 47450808,
      "step": 29505
    },
    {
      "epoch": 1.3359061997781752,
      "grad_norm": 0.4460858404636383,
      "learning_rate": 8.017009052400295e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 47458840,
      "step": 29510
    },
    {
      "epoch": 1.3361325516648181,
      "grad_norm": 0.2989516854286194,
      "learning_rate": 8.00980588008557e-06,
      "loss": 0.1181,
      "num_input_tokens_seen": 47466840,
      "step": 29515
    },
    {
      "epoch": 1.3363589035514611,
      "grad_norm": 0.7726073265075684,
      "learning_rate": 8.002605327872282e-06,
      "loss": 0.1383,
      "num_input_tokens_seen": 47474712,
      "step": 29520
    },
    {
      "epoch": 1.3365852554381041,
      "grad_norm": 0.3097488284111023,
      "learning_rate": 7.995407396870862e-06,
      "loss": 0.1186,
      "num_input_tokens_seen": 47482552,
      "step": 29525
    },
    {
      "epoch": 1.3368116073247471,
      "grad_norm": 0.84002685546875,
      "learning_rate": 7.988212088191307e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 47490744,
      "step": 29530
    },
    {
      "epoch": 1.33703795921139,
      "grad_norm": 0.28363102674484253,
      "learning_rate": 7.98101940294324e-06,
      "loss": 0.1166,
      "num_input_tokens_seen": 47498552,
      "step": 29535
    },
    {
      "epoch": 1.337264311098033,
      "grad_norm": 0.5227546095848083,
      "learning_rate": 7.973829342235847e-06,
      "loss": 0.1253,
      "num_input_tokens_seen": 47507192,
      "step": 29540
    },
    {
      "epoch": 1.337490662984676,
      "grad_norm": 1.284377932548523,
      "learning_rate": 7.966641907177936e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 47515512,
      "step": 29545
    },
    {
      "epoch": 1.3377170148713189,
      "grad_norm": 1.2589504718780518,
      "learning_rate": 7.959457098877901e-06,
      "loss": 0.1541,
      "num_input_tokens_seen": 47523512,
      "step": 29550
    },
    {
      "epoch": 1.3379433667579619,
      "grad_norm": 0.738839864730835,
      "learning_rate": 7.952274918443719e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 47531736,
      "step": 29555
    },
    {
      "epoch": 1.3381697186446049,
      "grad_norm": 0.5722161531448364,
      "learning_rate": 7.945095366982983e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 47539992,
      "step": 29560
    },
    {
      "epoch": 1.3383960705312479,
      "grad_norm": 0.7168716192245483,
      "learning_rate": 7.937918445602871e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 47548312,
      "step": 29565
    },
    {
      "epoch": 1.3386224224178909,
      "grad_norm": 0.4544956386089325,
      "learning_rate": 7.930744155410145e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 47555992,
      "step": 29570
    },
    {
      "epoch": 1.3388487743045339,
      "grad_norm": 1.0269478559494019,
      "learning_rate": 7.923572497511181e-06,
      "loss": 0.1513,
      "num_input_tokens_seen": 47563928,
      "step": 29575
    },
    {
      "epoch": 1.3390751261911769,
      "grad_norm": 1.0060193538665771,
      "learning_rate": 7.916403473011927e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 47572632,
      "step": 29580
    },
    {
      "epoch": 1.3393014780778199,
      "grad_norm": 0.5195742845535278,
      "learning_rate": 7.909237083017953e-06,
      "loss": 0.1346,
      "num_input_tokens_seen": 47580504,
      "step": 29585
    },
    {
      "epoch": 1.3395278299644628,
      "grad_norm": 0.964702308177948,
      "learning_rate": 7.902073328634389e-06,
      "loss": 0.159,
      "num_input_tokens_seen": 47589560,
      "step": 29590
    },
    {
      "epoch": 1.3397541818511058,
      "grad_norm": 1.0267139673233032,
      "learning_rate": 7.894912210965987e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 47597432,
      "step": 29595
    },
    {
      "epoch": 1.3399805337377488,
      "grad_norm": 0.68062424659729,
      "learning_rate": 7.887753731117075e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 47605208,
      "step": 29600
    },
    {
      "epoch": 1.3399805337377488,
      "eval_loss": 0.14422903954982758,
      "eval_runtime": 405.4066,
      "eval_samples_per_second": 96.868,
      "eval_steps_per_second": 24.218,
      "num_input_tokens_seen": 47605208,
      "step": 29600
    },
    {
      "epoch": 1.3402068856243916,
      "grad_norm": 1.1560344696044922,
      "learning_rate": 7.880597890191587e-06,
      "loss": 0.1487,
      "num_input_tokens_seen": 47613336,
      "step": 29605
    },
    {
      "epoch": 1.3404332375110346,
      "grad_norm": 0.9334280490875244,
      "learning_rate": 7.873444689293036e-06,
      "loss": 0.1236,
      "num_input_tokens_seen": 47621112,
      "step": 29610
    },
    {
      "epoch": 1.3406595893976776,
      "grad_norm": 0.8684763312339783,
      "learning_rate": 7.866294129524548e-06,
      "loss": 0.1574,
      "num_input_tokens_seen": 47629048,
      "step": 29615
    },
    {
      "epoch": 1.3408859412843206,
      "grad_norm": 0.7784403562545776,
      "learning_rate": 7.859146211988811e-06,
      "loss": 0.1765,
      "num_input_tokens_seen": 47637976,
      "step": 29620
    },
    {
      "epoch": 1.3411122931709636,
      "grad_norm": 0.6116704940795898,
      "learning_rate": 7.852000937788134e-06,
      "loss": 0.1938,
      "num_input_tokens_seen": 47645816,
      "step": 29625
    },
    {
      "epoch": 1.3413386450576066,
      "grad_norm": 0.5532876253128052,
      "learning_rate": 7.844858308024416e-06,
      "loss": 0.1186,
      "num_input_tokens_seen": 47653880,
      "step": 29630
    },
    {
      "epoch": 1.3415649969442496,
      "grad_norm": 0.5606434345245361,
      "learning_rate": 7.837718323799122e-06,
      "loss": 0.1091,
      "num_input_tokens_seen": 47661592,
      "step": 29635
    },
    {
      "epoch": 1.3417913488308926,
      "grad_norm": 0.3419717252254486,
      "learning_rate": 7.83058098621334e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 47669752,
      "step": 29640
    },
    {
      "epoch": 1.3420177007175353,
      "grad_norm": 0.1858600527048111,
      "learning_rate": 7.823446296367739e-06,
      "loss": 0.1313,
      "num_input_tokens_seen": 47677496,
      "step": 29645
    },
    {
      "epoch": 1.3422440526041783,
      "grad_norm": 0.5991155505180359,
      "learning_rate": 7.81631425536257e-06,
      "loss": 0.1379,
      "num_input_tokens_seen": 47685752,
      "step": 29650
    },
    {
      "epoch": 1.3424704044908213,
      "grad_norm": 0.5397133231163025,
      "learning_rate": 7.809184864297689e-06,
      "loss": 0.171,
      "num_input_tokens_seen": 47694072,
      "step": 29655
    },
    {
      "epoch": 1.3426967563774643,
      "grad_norm": 0.7718371152877808,
      "learning_rate": 7.802058124272532e-06,
      "loss": 0.1652,
      "num_input_tokens_seen": 47701912,
      "step": 29660
    },
    {
      "epoch": 1.3429231082641073,
      "grad_norm": 0.303477942943573,
      "learning_rate": 7.79493403638614e-06,
      "loss": 0.0745,
      "num_input_tokens_seen": 47710040,
      "step": 29665
    },
    {
      "epoch": 1.3431494601507503,
      "grad_norm": 0.7113016247749329,
      "learning_rate": 7.787812601737132e-06,
      "loss": 0.13,
      "num_input_tokens_seen": 47718104,
      "step": 29670
    },
    {
      "epoch": 1.3433758120373933,
      "grad_norm": 0.5667906403541565,
      "learning_rate": 7.780693821423715e-06,
      "loss": 0.1167,
      "num_input_tokens_seen": 47726200,
      "step": 29675
    },
    {
      "epoch": 1.3436021639240363,
      "grad_norm": 0.8417982459068298,
      "learning_rate": 7.773577696543705e-06,
      "loss": 0.1273,
      "num_input_tokens_seen": 47734072,
      "step": 29680
    },
    {
      "epoch": 1.3438285158106793,
      "grad_norm": 0.7102984189987183,
      "learning_rate": 7.7664642281945e-06,
      "loss": 0.1496,
      "num_input_tokens_seen": 47741912,
      "step": 29685
    },
    {
      "epoch": 1.3440548676973223,
      "grad_norm": 0.5499508380889893,
      "learning_rate": 7.759353417473072e-06,
      "loss": 0.098,
      "num_input_tokens_seen": 47749912,
      "step": 29690
    },
    {
      "epoch": 1.3442812195839653,
      "grad_norm": 0.6815902590751648,
      "learning_rate": 7.752245265476016e-06,
      "loss": 0.1282,
      "num_input_tokens_seen": 47758104,
      "step": 29695
    },
    {
      "epoch": 1.3445075714706083,
      "grad_norm": 0.8217626810073853,
      "learning_rate": 7.745139773299481e-06,
      "loss": 0.1087,
      "num_input_tokens_seen": 47766296,
      "step": 29700
    },
    {
      "epoch": 1.3447339233572513,
      "grad_norm": 0.9049551486968994,
      "learning_rate": 7.738036942039232e-06,
      "loss": 0.1433,
      "num_input_tokens_seen": 47774040,
      "step": 29705
    },
    {
      "epoch": 1.3449602752438943,
      "grad_norm": 0.9854346513748169,
      "learning_rate": 7.73093677279062e-06,
      "loss": 0.1376,
      "num_input_tokens_seen": 47782136,
      "step": 29710
    },
    {
      "epoch": 1.3451866271305373,
      "grad_norm": 1.011861801147461,
      "learning_rate": 7.72383926664857e-06,
      "loss": 0.175,
      "num_input_tokens_seen": 47790104,
      "step": 29715
    },
    {
      "epoch": 1.34541297901718,
      "grad_norm": 0.8216250538825989,
      "learning_rate": 7.716744424707606e-06,
      "loss": 0.1714,
      "num_input_tokens_seen": 47797496,
      "step": 29720
    },
    {
      "epoch": 1.345639330903823,
      "grad_norm": 0.5417309403419495,
      "learning_rate": 7.709652248061858e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 47805656,
      "step": 29725
    },
    {
      "epoch": 1.345865682790466,
      "grad_norm": 0.6536298990249634,
      "learning_rate": 7.702562737805017e-06,
      "loss": 0.1661,
      "num_input_tokens_seen": 47813656,
      "step": 29730
    },
    {
      "epoch": 1.346092034677109,
      "grad_norm": 0.9168511629104614,
      "learning_rate": 7.695475895030365e-06,
      "loss": 0.1516,
      "num_input_tokens_seen": 47821304,
      "step": 29735
    },
    {
      "epoch": 1.346318386563752,
      "grad_norm": 0.7182347774505615,
      "learning_rate": 7.6883917208308e-06,
      "loss": 0.1541,
      "num_input_tokens_seen": 47829400,
      "step": 29740
    },
    {
      "epoch": 1.346544738450395,
      "grad_norm": 0.5953862071037292,
      "learning_rate": 7.681310216298778e-06,
      "loss": 0.1582,
      "num_input_tokens_seen": 47837112,
      "step": 29745
    },
    {
      "epoch": 1.346771090337038,
      "grad_norm": 0.9476062059402466,
      "learning_rate": 7.674231382526367e-06,
      "loss": 0.1688,
      "num_input_tokens_seen": 47844920,
      "step": 29750
    },
    {
      "epoch": 1.346997442223681,
      "grad_norm": 1.314544677734375,
      "learning_rate": 7.667155220605198e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 47853272,
      "step": 29755
    },
    {
      "epoch": 1.3472237941103238,
      "grad_norm": 0.5506774187088013,
      "learning_rate": 7.660081731626515e-06,
      "loss": 0.1153,
      "num_input_tokens_seen": 47861656,
      "step": 29760
    },
    {
      "epoch": 1.3474501459969668,
      "grad_norm": 0.6923278570175171,
      "learning_rate": 7.653010916681141e-06,
      "loss": 0.1131,
      "num_input_tokens_seen": 47869176,
      "step": 29765
    },
    {
      "epoch": 1.3476764978836098,
      "grad_norm": 0.3455384373664856,
      "learning_rate": 7.645942776859472e-06,
      "loss": 0.1225,
      "num_input_tokens_seen": 47877208,
      "step": 29770
    },
    {
      "epoch": 1.3479028497702528,
      "grad_norm": 0.5116740465164185,
      "learning_rate": 7.63887731325152e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 47885240,
      "step": 29775
    },
    {
      "epoch": 1.3481292016568958,
      "grad_norm": 0.7124277949333191,
      "learning_rate": 7.63181452694685e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 47893048,
      "step": 29780
    },
    {
      "epoch": 1.3483555535435388,
      "grad_norm": 0.8060470223426819,
      "learning_rate": 7.624754419034644e-06,
      "loss": 0.1661,
      "num_input_tokens_seen": 47900792,
      "step": 29785
    },
    {
      "epoch": 1.3485819054301817,
      "grad_norm": 0.6199606657028198,
      "learning_rate": 7.6176969906036645e-06,
      "loss": 0.1052,
      "num_input_tokens_seen": 47909048,
      "step": 29790
    },
    {
      "epoch": 1.3488082573168247,
      "grad_norm": 0.6440979242324829,
      "learning_rate": 7.610642242742242e-06,
      "loss": 0.1488,
      "num_input_tokens_seen": 47917784,
      "step": 29795
    },
    {
      "epoch": 1.3490346092034677,
      "grad_norm": 1.3884971141815186,
      "learning_rate": 7.603590176538322e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 47925944,
      "step": 29800
    },
    {
      "epoch": 1.3490346092034677,
      "eval_loss": 0.14380232989788055,
      "eval_runtime": 405.2599,
      "eval_samples_per_second": 96.903,
      "eval_steps_per_second": 24.226,
      "num_input_tokens_seen": 47925944,
      "step": 29800
    },
    {
      "epoch": 1.3492609610901107,
      "grad_norm": 0.4682426154613495,
      "learning_rate": 7.596540793079404e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 47933976,
      "step": 29805
    },
    {
      "epoch": 1.3494873129767537,
      "grad_norm": 0.9042382836341858,
      "learning_rate": 7.5894940934526125e-06,
      "loss": 0.1933,
      "num_input_tokens_seen": 47941592,
      "step": 29810
    },
    {
      "epoch": 1.3497136648633967,
      "grad_norm": 0.7287344336509705,
      "learning_rate": 7.582450078744621e-06,
      "loss": 0.1714,
      "num_input_tokens_seen": 47949816,
      "step": 29815
    },
    {
      "epoch": 1.3499400167500397,
      "grad_norm": 0.9938543438911438,
      "learning_rate": 7.575408750041707e-06,
      "loss": 0.1649,
      "num_input_tokens_seen": 47958424,
      "step": 29820
    },
    {
      "epoch": 1.3501663686366827,
      "grad_norm": 0.5139813423156738,
      "learning_rate": 7.568370108429732e-06,
      "loss": 0.1367,
      "num_input_tokens_seen": 47966328,
      "step": 29825
    },
    {
      "epoch": 1.3503927205233255,
      "grad_norm": 1.1348460912704468,
      "learning_rate": 7.561334154994154e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 47974936,
      "step": 29830
    },
    {
      "epoch": 1.3506190724099685,
      "grad_norm": 0.4678806662559509,
      "learning_rate": 7.55430089081999e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 47983000,
      "step": 29835
    },
    {
      "epoch": 1.3508454242966115,
      "grad_norm": 0.877358615398407,
      "learning_rate": 7.547270316991864e-06,
      "loss": 0.1497,
      "num_input_tokens_seen": 47991160,
      "step": 29840
    },
    {
      "epoch": 1.3510717761832545,
      "grad_norm": 0.4724838137626648,
      "learning_rate": 7.5402424345939884e-06,
      "loss": 0.1079,
      "num_input_tokens_seen": 47999224,
      "step": 29845
    },
    {
      "epoch": 1.3512981280698975,
      "grad_norm": 0.49899426102638245,
      "learning_rate": 7.533217244710133e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 48007192,
      "step": 29850
    },
    {
      "epoch": 1.3515244799565405,
      "grad_norm": 0.3271738588809967,
      "learning_rate": 7.52619474842369e-06,
      "loss": 0.148,
      "num_input_tokens_seen": 48015960,
      "step": 29855
    },
    {
      "epoch": 1.3517508318431835,
      "grad_norm": 0.42531684041023254,
      "learning_rate": 7.519174946817597e-06,
      "loss": 0.1536,
      "num_input_tokens_seen": 48024408,
      "step": 29860
    },
    {
      "epoch": 1.3519771837298264,
      "grad_norm": 0.3496760129928589,
      "learning_rate": 7.512157840974407e-06,
      "loss": 0.1167,
      "num_input_tokens_seen": 48032408,
      "step": 29865
    },
    {
      "epoch": 1.3522035356164694,
      "grad_norm": 0.9117850661277771,
      "learning_rate": 7.5051434319762496e-06,
      "loss": 0.1277,
      "num_input_tokens_seen": 48040312,
      "step": 29870
    },
    {
      "epoch": 1.3524298875031122,
      "grad_norm": 0.5475383400917053,
      "learning_rate": 7.498131720904822e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 48048664,
      "step": 29875
    },
    {
      "epoch": 1.3526562393897552,
      "grad_norm": 0.40795090794563293,
      "learning_rate": 7.491122708841433e-06,
      "loss": 0.1698,
      "num_input_tokens_seen": 48057048,
      "step": 29880
    },
    {
      "epoch": 1.3528825912763982,
      "grad_norm": 0.6499723792076111,
      "learning_rate": 7.4841163968669524e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 48065560,
      "step": 29885
    },
    {
      "epoch": 1.3531089431630412,
      "grad_norm": 0.33784398436546326,
      "learning_rate": 7.4771127860618355e-06,
      "loss": 0.126,
      "num_input_tokens_seen": 48073336,
      "step": 29890
    },
    {
      "epoch": 1.3533352950496842,
      "grad_norm": 0.5268245339393616,
      "learning_rate": 7.470111877506139e-06,
      "loss": 0.1402,
      "num_input_tokens_seen": 48081240,
      "step": 29895
    },
    {
      "epoch": 1.3535616469363272,
      "grad_norm": 0.8234447240829468,
      "learning_rate": 7.463113672279479e-06,
      "loss": 0.1447,
      "num_input_tokens_seen": 48089528,
      "step": 29900
    },
    {
      "epoch": 1.3537879988229702,
      "grad_norm": 0.4310288429260254,
      "learning_rate": 7.456118171461071e-06,
      "loss": 0.154,
      "num_input_tokens_seen": 48097240,
      "step": 29905
    },
    {
      "epoch": 1.3540143507096132,
      "grad_norm": 0.7223504185676575,
      "learning_rate": 7.449125376129721e-06,
      "loss": 0.1533,
      "num_input_tokens_seen": 48104664,
      "step": 29910
    },
    {
      "epoch": 1.3542407025962562,
      "grad_norm": 0.33356690406799316,
      "learning_rate": 7.442135287363788e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 48113368,
      "step": 29915
    },
    {
      "epoch": 1.3544670544828992,
      "grad_norm": 0.2820444107055664,
      "learning_rate": 7.435147906241247e-06,
      "loss": 0.1268,
      "num_input_tokens_seen": 48121336,
      "step": 29920
    },
    {
      "epoch": 1.3546934063695422,
      "grad_norm": 0.7533041834831238,
      "learning_rate": 7.428163233839624e-06,
      "loss": 0.1724,
      "num_input_tokens_seen": 48130264,
      "step": 29925
    },
    {
      "epoch": 1.3549197582561852,
      "grad_norm": 0.3641285300254822,
      "learning_rate": 7.4211812712360525e-06,
      "loss": 0.1313,
      "num_input_tokens_seen": 48138072,
      "step": 29930
    },
    {
      "epoch": 1.3551461101428282,
      "grad_norm": 0.38067036867141724,
      "learning_rate": 7.4142020195072464e-06,
      "loss": 0.1126,
      "num_input_tokens_seen": 48146104,
      "step": 29935
    },
    {
      "epoch": 1.3553724620294711,
      "grad_norm": 0.866533100605011,
      "learning_rate": 7.407225479729479e-06,
      "loss": 0.1373,
      "num_input_tokens_seen": 48155000,
      "step": 29940
    },
    {
      "epoch": 1.355598813916114,
      "grad_norm": 0.5417168140411377,
      "learning_rate": 7.400251652978632e-06,
      "loss": 0.1225,
      "num_input_tokens_seen": 48162616,
      "step": 29945
    },
    {
      "epoch": 1.355825165802757,
      "grad_norm": 0.6555594205856323,
      "learning_rate": 7.393280540330147e-06,
      "loss": 0.1767,
      "num_input_tokens_seen": 48171064,
      "step": 29950
    },
    {
      "epoch": 1.3560515176894,
      "grad_norm": 0.8381404876708984,
      "learning_rate": 7.386312142859069e-06,
      "loss": 0.1439,
      "num_input_tokens_seen": 48178840,
      "step": 29955
    },
    {
      "epoch": 1.356277869576043,
      "grad_norm": 0.2453487068414688,
      "learning_rate": 7.379346461640008e-06,
      "loss": 0.1254,
      "num_input_tokens_seen": 48186424,
      "step": 29960
    },
    {
      "epoch": 1.356504221462686,
      "grad_norm": 0.946552574634552,
      "learning_rate": 7.372383497747149e-06,
      "loss": 0.1621,
      "num_input_tokens_seen": 48194072,
      "step": 29965
    },
    {
      "epoch": 1.356730573349329,
      "grad_norm": 0.7560420036315918,
      "learning_rate": 7.3654232522542775e-06,
      "loss": 0.1301,
      "num_input_tokens_seen": 48202712,
      "step": 29970
    },
    {
      "epoch": 1.356956925235972,
      "grad_norm": 1.1533557176589966,
      "learning_rate": 7.358465726234756e-06,
      "loss": 0.1786,
      "num_input_tokens_seen": 48210552,
      "step": 29975
    },
    {
      "epoch": 1.3571832771226149,
      "grad_norm": 0.995723307132721,
      "learning_rate": 7.351510920761512e-06,
      "loss": 0.0996,
      "num_input_tokens_seen": 48218584,
      "step": 29980
    },
    {
      "epoch": 1.3574096290092577,
      "grad_norm": 1.1621787548065186,
      "learning_rate": 7.344558836907067e-06,
      "loss": 0.1669,
      "num_input_tokens_seen": 48226424,
      "step": 29985
    },
    {
      "epoch": 1.3576359808959007,
      "grad_norm": 0.24782608449459076,
      "learning_rate": 7.3376094757435285e-06,
      "loss": 0.1095,
      "num_input_tokens_seen": 48234392,
      "step": 29990
    },
    {
      "epoch": 1.3578623327825436,
      "grad_norm": 0.6221591830253601,
      "learning_rate": 7.330662838342561e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 48242200,
      "step": 29995
    },
    {
      "epoch": 1.3580886846691866,
      "grad_norm": 2.0678489208221436,
      "learning_rate": 7.323718925775438e-06,
      "loss": 0.1625,
      "num_input_tokens_seen": 48250168,
      "step": 30000
    },
    {
      "epoch": 1.3580886846691866,
      "eval_loss": 0.14424416422843933,
      "eval_runtime": 405.5897,
      "eval_samples_per_second": 96.824,
      "eval_steps_per_second": 24.207,
      "num_input_tokens_seen": 48250168,
      "step": 30000
    },
    {
      "epoch": 1.3583150365558296,
      "grad_norm": 0.8434796333312988,
      "learning_rate": 7.316777739112985e-06,
      "loss": 0.1564,
      "num_input_tokens_seen": 48257944,
      "step": 30005
    },
    {
      "epoch": 1.3585413884424726,
      "grad_norm": 0.818418025970459,
      "learning_rate": 7.309839279425626e-06,
      "loss": 0.1367,
      "num_input_tokens_seen": 48265944,
      "step": 30010
    },
    {
      "epoch": 1.3587677403291156,
      "grad_norm": 0.5295703411102295,
      "learning_rate": 7.302903547783366e-06,
      "loss": 0.1201,
      "num_input_tokens_seen": 48273880,
      "step": 30015
    },
    {
      "epoch": 1.3589940922157586,
      "grad_norm": 0.8145907521247864,
      "learning_rate": 7.2959705452557644e-06,
      "loss": 0.114,
      "num_input_tokens_seen": 48282040,
      "step": 30020
    },
    {
      "epoch": 1.3592204441024016,
      "grad_norm": 0.39770692586898804,
      "learning_rate": 7.289040272911996e-06,
      "loss": 0.118,
      "num_input_tokens_seen": 48289560,
      "step": 30025
    },
    {
      "epoch": 1.3594467959890446,
      "grad_norm": 1.5233832597732544,
      "learning_rate": 7.282112731820789e-06,
      "loss": 0.152,
      "num_input_tokens_seen": 48297272,
      "step": 30030
    },
    {
      "epoch": 1.3596731478756876,
      "grad_norm": 0.8273726105690002,
      "learning_rate": 7.275187923050447e-06,
      "loss": 0.1246,
      "num_input_tokens_seen": 48305176,
      "step": 30035
    },
    {
      "epoch": 1.3598994997623306,
      "grad_norm": 1.200165033340454,
      "learning_rate": 7.268265847668879e-06,
      "loss": 0.1501,
      "num_input_tokens_seen": 48313272,
      "step": 30040
    },
    {
      "epoch": 1.3601258516489736,
      "grad_norm": 1.1675376892089844,
      "learning_rate": 7.261346506743538e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 48321368,
      "step": 30045
    },
    {
      "epoch": 1.3603522035356166,
      "grad_norm": 1.1142266988754272,
      "learning_rate": 7.254429901341486e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 48328952,
      "step": 30050
    },
    {
      "epoch": 1.3605785554222594,
      "grad_norm": 1.1358110904693604,
      "learning_rate": 7.247516032529356e-06,
      "loss": 0.1651,
      "num_input_tokens_seen": 48337432,
      "step": 30055
    },
    {
      "epoch": 1.3608049073089024,
      "grad_norm": 1.0356645584106445,
      "learning_rate": 7.240604901373338e-06,
      "loss": 0.1673,
      "num_input_tokens_seen": 48345752,
      "step": 30060
    },
    {
      "epoch": 1.3610312591955454,
      "grad_norm": 0.44950199127197266,
      "learning_rate": 7.233696508939223e-06,
      "loss": 0.1146,
      "num_input_tokens_seen": 48353656,
      "step": 30065
    },
    {
      "epoch": 1.3612576110821883,
      "grad_norm": 0.5226582884788513,
      "learning_rate": 7.226790856292376e-06,
      "loss": 0.1211,
      "num_input_tokens_seen": 48361176,
      "step": 30070
    },
    {
      "epoch": 1.3614839629688313,
      "grad_norm": 0.5920550227165222,
      "learning_rate": 7.219887944497727e-06,
      "loss": 0.1248,
      "num_input_tokens_seen": 48368856,
      "step": 30075
    },
    {
      "epoch": 1.3617103148554743,
      "grad_norm": 0.9927315711975098,
      "learning_rate": 7.2129877746198e-06,
      "loss": 0.1976,
      "num_input_tokens_seen": 48376760,
      "step": 30080
    },
    {
      "epoch": 1.3619366667421173,
      "grad_norm": 1.0745654106140137,
      "learning_rate": 7.20609034772268e-06,
      "loss": 0.1198,
      "num_input_tokens_seen": 48385016,
      "step": 30085
    },
    {
      "epoch": 1.3621630186287603,
      "grad_norm": 0.9372663497924805,
      "learning_rate": 7.19919566487004e-06,
      "loss": 0.1232,
      "num_input_tokens_seen": 48393464,
      "step": 30090
    },
    {
      "epoch": 1.3623893705154033,
      "grad_norm": 0.42604222893714905,
      "learning_rate": 7.192303727125132e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 48401336,
      "step": 30095
    },
    {
      "epoch": 1.362615722402046,
      "grad_norm": 1.10054612159729,
      "learning_rate": 7.185414535550777e-06,
      "loss": 0.1611,
      "num_input_tokens_seen": 48409912,
      "step": 30100
    },
    {
      "epoch": 1.362842074288689,
      "grad_norm": 0.5739171504974365,
      "learning_rate": 7.178528091209363e-06,
      "loss": 0.1367,
      "num_input_tokens_seen": 48417400,
      "step": 30105
    },
    {
      "epoch": 1.363068426175332,
      "grad_norm": 1.1953812837600708,
      "learning_rate": 7.171644395162888e-06,
      "loss": 0.1654,
      "num_input_tokens_seen": 48425048,
      "step": 30110
    },
    {
      "epoch": 1.363294778061975,
      "grad_norm": 0.3378751575946808,
      "learning_rate": 7.164763448472881e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 48432920,
      "step": 30115
    },
    {
      "epoch": 1.363521129948618,
      "grad_norm": 0.6244768500328064,
      "learning_rate": 7.157885252200491e-06,
      "loss": 0.1178,
      "num_input_tokens_seen": 48440888,
      "step": 30120
    },
    {
      "epoch": 1.363747481835261,
      "grad_norm": 0.5955641865730286,
      "learning_rate": 7.151009807406403e-06,
      "loss": 0.1175,
      "num_input_tokens_seen": 48449144,
      "step": 30125
    },
    {
      "epoch": 1.363973833721904,
      "grad_norm": 0.9485923647880554,
      "learning_rate": 7.144137115150909e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 48456696,
      "step": 30130
    },
    {
      "epoch": 1.364200185608547,
      "grad_norm": 0.6335325241088867,
      "learning_rate": 7.1372671764938725e-06,
      "loss": 0.151,
      "num_input_tokens_seen": 48464888,
      "step": 30135
    },
    {
      "epoch": 1.36442653749519,
      "grad_norm": 0.968186616897583,
      "learning_rate": 7.130399992494705e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 48472568,
      "step": 30140
    },
    {
      "epoch": 1.364652889381833,
      "grad_norm": 0.43047407269477844,
      "learning_rate": 7.123535564212419e-06,
      "loss": 0.1123,
      "num_input_tokens_seen": 48480536,
      "step": 30145
    },
    {
      "epoch": 1.364879241268476,
      "grad_norm": 0.44126084446907043,
      "learning_rate": 7.116673892705611e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 48488504,
      "step": 30150
    },
    {
      "epoch": 1.365105593155119,
      "grad_norm": 0.2631925046443939,
      "learning_rate": 7.109814979032415e-06,
      "loss": 0.1634,
      "num_input_tokens_seen": 48496440,
      "step": 30155
    },
    {
      "epoch": 1.365331945041762,
      "grad_norm": 0.7708076238632202,
      "learning_rate": 7.102958824250577e-06,
      "loss": 0.1532,
      "num_input_tokens_seen": 48504120,
      "step": 30160
    },
    {
      "epoch": 1.365558296928405,
      "grad_norm": 0.49119672179222107,
      "learning_rate": 7.096105429417393e-06,
      "loss": 0.1653,
      "num_input_tokens_seen": 48511896,
      "step": 30165
    },
    {
      "epoch": 1.3657846488150478,
      "grad_norm": 0.3475629985332489,
      "learning_rate": 7.0892547955897506e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 48520024,
      "step": 30170
    },
    {
      "epoch": 1.3660110007016908,
      "grad_norm": 0.6473761796951294,
      "learning_rate": 7.0824069238241e-06,
      "loss": 0.1285,
      "num_input_tokens_seen": 48528152,
      "step": 30175
    },
    {
      "epoch": 1.3662373525883338,
      "grad_norm": 0.7164068818092346,
      "learning_rate": 7.075561815176462e-06,
      "loss": 0.1113,
      "num_input_tokens_seen": 48536248,
      "step": 30180
    },
    {
      "epoch": 1.3664637044749768,
      "grad_norm": 0.4566408097743988,
      "learning_rate": 7.068719470702445e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 48544152,
      "step": 30185
    },
    {
      "epoch": 1.3666900563616198,
      "grad_norm": 0.5394814610481262,
      "learning_rate": 7.061879891457229e-06,
      "loss": 0.1247,
      "num_input_tokens_seen": 48553560,
      "step": 30190
    },
    {
      "epoch": 1.3669164082482628,
      "grad_norm": 0.5342133045196533,
      "learning_rate": 7.0550430784955515e-06,
      "loss": 0.1349,
      "num_input_tokens_seen": 48561720,
      "step": 30195
    },
    {
      "epoch": 1.3671427601349058,
      "grad_norm": 0.7475462555885315,
      "learning_rate": 7.048209032871752e-06,
      "loss": 0.1181,
      "num_input_tokens_seen": 48570040,
      "step": 30200
    },
    {
      "epoch": 1.3671427601349058,
      "eval_loss": 0.14372391998767853,
      "eval_runtime": 405.4036,
      "eval_samples_per_second": 96.869,
      "eval_steps_per_second": 24.218,
      "num_input_tokens_seen": 48570040,
      "step": 30200
    },
    {
      "epoch": 1.3673691120215488,
      "grad_norm": 0.5065116882324219,
      "learning_rate": 7.0413777556397055e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 48578488,
      "step": 30205
    },
    {
      "epoch": 1.3675954639081918,
      "grad_norm": 0.3723568022251129,
      "learning_rate": 7.0345492478528925e-06,
      "loss": 0.1201,
      "num_input_tokens_seen": 48585976,
      "step": 30210
    },
    {
      "epoch": 1.3678218157948345,
      "grad_norm": 0.4070613980293274,
      "learning_rate": 7.02772351056436e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 48593848,
      "step": 30215
    },
    {
      "epoch": 1.3680481676814775,
      "grad_norm": 1.1464122533798218,
      "learning_rate": 7.020900544826709e-06,
      "loss": 0.1341,
      "num_input_tokens_seen": 48601880,
      "step": 30220
    },
    {
      "epoch": 1.3682745195681205,
      "grad_norm": 0.6085906028747559,
      "learning_rate": 7.014080351692134e-06,
      "loss": 0.1355,
      "num_input_tokens_seen": 48609528,
      "step": 30225
    },
    {
      "epoch": 1.3685008714547635,
      "grad_norm": 0.5565903782844543,
      "learning_rate": 7.0072629322124024e-06,
      "loss": 0.1451,
      "num_input_tokens_seen": 48617496,
      "step": 30230
    },
    {
      "epoch": 1.3687272233414065,
      "grad_norm": 0.27795860171318054,
      "learning_rate": 7.000448287438827e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 48625944,
      "step": 30235
    },
    {
      "epoch": 1.3689535752280495,
      "grad_norm": 0.6696791648864746,
      "learning_rate": 6.993636418422331e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 48634008,
      "step": 30240
    },
    {
      "epoch": 1.3691799271146925,
      "grad_norm": 0.7867993116378784,
      "learning_rate": 6.986827326213383e-06,
      "loss": 0.1653,
      "num_input_tokens_seen": 48642488,
      "step": 30245
    },
    {
      "epoch": 1.3694062790013355,
      "grad_norm": 1.3831884860992432,
      "learning_rate": 6.9800210118620205e-06,
      "loss": 0.1664,
      "num_input_tokens_seen": 48650552,
      "step": 30250
    },
    {
      "epoch": 1.3696326308879785,
      "grad_norm": 0.5969195365905762,
      "learning_rate": 6.973217476417876e-06,
      "loss": 0.151,
      "num_input_tokens_seen": 48658040,
      "step": 30255
    },
    {
      "epoch": 1.3698589827746215,
      "grad_norm": 1.0479137897491455,
      "learning_rate": 6.96641672093013e-06,
      "loss": 0.1479,
      "num_input_tokens_seen": 48666424,
      "step": 30260
    },
    {
      "epoch": 1.3700853346612645,
      "grad_norm": 0.6355612277984619,
      "learning_rate": 6.95961874644755e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 48674584,
      "step": 30265
    },
    {
      "epoch": 1.3703116865479075,
      "grad_norm": 0.5951977968215942,
      "learning_rate": 6.952823554018476e-06,
      "loss": 0.1429,
      "num_input_tokens_seen": 48683448,
      "step": 30270
    },
    {
      "epoch": 1.3705380384345505,
      "grad_norm": 0.9241906404495239,
      "learning_rate": 6.946031144690798e-06,
      "loss": 0.1426,
      "num_input_tokens_seen": 48691160,
      "step": 30275
    },
    {
      "epoch": 1.3707643903211935,
      "grad_norm": 0.6267548203468323,
      "learning_rate": 6.939241519512005e-06,
      "loss": 0.148,
      "num_input_tokens_seen": 48699000,
      "step": 30280
    },
    {
      "epoch": 1.3709907422078362,
      "grad_norm": 0.8574019074440002,
      "learning_rate": 6.932454679529129e-06,
      "loss": 0.1232,
      "num_input_tokens_seen": 48706616,
      "step": 30285
    },
    {
      "epoch": 1.3712170940944792,
      "grad_norm": 1.2207268476486206,
      "learning_rate": 6.925670625788791e-06,
      "loss": 0.1824,
      "num_input_tokens_seen": 48714488,
      "step": 30290
    },
    {
      "epoch": 1.3714434459811222,
      "grad_norm": 1.0004054307937622,
      "learning_rate": 6.918889359337186e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 48722136,
      "step": 30295
    },
    {
      "epoch": 1.3716697978677652,
      "grad_norm": 0.35358279943466187,
      "learning_rate": 6.912110881220058e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 48730264,
      "step": 30300
    },
    {
      "epoch": 1.3718961497544082,
      "grad_norm": 0.6952992677688599,
      "learning_rate": 6.905335192482735e-06,
      "loss": 0.1658,
      "num_input_tokens_seen": 48738392,
      "step": 30305
    },
    {
      "epoch": 1.3721225016410512,
      "grad_norm": 1.252199411392212,
      "learning_rate": 6.8985622941701275e-06,
      "loss": 0.1125,
      "num_input_tokens_seen": 48745880,
      "step": 30310
    },
    {
      "epoch": 1.3723488535276942,
      "grad_norm": 0.5409765839576721,
      "learning_rate": 6.89179218732669e-06,
      "loss": 0.1655,
      "num_input_tokens_seen": 48754008,
      "step": 30315
    },
    {
      "epoch": 1.3725752054143372,
      "grad_norm": 0.8085992932319641,
      "learning_rate": 6.8850248729964595e-06,
      "loss": 0.1629,
      "num_input_tokens_seen": 48762136,
      "step": 30320
    },
    {
      "epoch": 1.37280155730098,
      "grad_norm": 0.3401549160480499,
      "learning_rate": 6.8782603522230314e-06,
      "loss": 0.181,
      "num_input_tokens_seen": 48770456,
      "step": 30325
    },
    {
      "epoch": 1.373027909187623,
      "grad_norm": 0.6845285892486572,
      "learning_rate": 6.871498626049591e-06,
      "loss": 0.1086,
      "num_input_tokens_seen": 48778168,
      "step": 30330
    },
    {
      "epoch": 1.373254261074266,
      "grad_norm": 0.24924878776073456,
      "learning_rate": 6.8647396955188875e-06,
      "loss": 0.1506,
      "num_input_tokens_seen": 48786200,
      "step": 30335
    },
    {
      "epoch": 1.373480612960909,
      "grad_norm": 0.5676418542861938,
      "learning_rate": 6.857983561673218e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 48793688,
      "step": 30340
    },
    {
      "epoch": 1.373706964847552,
      "grad_norm": 0.5504829287528992,
      "learning_rate": 6.851230225554467e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 48801528,
      "step": 30345
    },
    {
      "epoch": 1.373933316734195,
      "grad_norm": 0.7531188130378723,
      "learning_rate": 6.8444796882040946e-06,
      "loss": 0.1559,
      "num_input_tokens_seen": 48809848,
      "step": 30350
    },
    {
      "epoch": 1.374159668620838,
      "grad_norm": 0.4539324641227722,
      "learning_rate": 6.837731950663106e-06,
      "loss": 0.1199,
      "num_input_tokens_seen": 48817784,
      "step": 30355
    },
    {
      "epoch": 1.374386020507481,
      "grad_norm": 0.7775808572769165,
      "learning_rate": 6.830987013972098e-06,
      "loss": 0.1766,
      "num_input_tokens_seen": 48826264,
      "step": 30360
    },
    {
      "epoch": 1.374612372394124,
      "grad_norm": 0.5042804479598999,
      "learning_rate": 6.82424487917121e-06,
      "loss": 0.1473,
      "num_input_tokens_seen": 48834360,
      "step": 30365
    },
    {
      "epoch": 1.374838724280767,
      "grad_norm": 1.02259361743927,
      "learning_rate": 6.8175055473001735e-06,
      "loss": 0.1452,
      "num_input_tokens_seen": 48843448,
      "step": 30370
    },
    {
      "epoch": 1.37506507616741,
      "grad_norm": 0.5416017770767212,
      "learning_rate": 6.8107690193982855e-06,
      "loss": 0.1449,
      "num_input_tokens_seen": 48851256,
      "step": 30375
    },
    {
      "epoch": 1.375291428054053,
      "grad_norm": 0.7041333913803101,
      "learning_rate": 6.804035296504385e-06,
      "loss": 0.168,
      "num_input_tokens_seen": 48859416,
      "step": 30380
    },
    {
      "epoch": 1.375517779940696,
      "grad_norm": 0.521588146686554,
      "learning_rate": 6.797304379656916e-06,
      "loss": 0.1739,
      "num_input_tokens_seen": 48867864,
      "step": 30385
    },
    {
      "epoch": 1.375744131827339,
      "grad_norm": 0.35544678568840027,
      "learning_rate": 6.790576269893861e-06,
      "loss": 0.1286,
      "num_input_tokens_seen": 48875736,
      "step": 30390
    },
    {
      "epoch": 1.3759704837139817,
      "grad_norm": 0.4527750015258789,
      "learning_rate": 6.783850968252772e-06,
      "loss": 0.1998,
      "num_input_tokens_seen": 48883736,
      "step": 30395
    },
    {
      "epoch": 1.3761968356006247,
      "grad_norm": 0.4467397928237915,
      "learning_rate": 6.777128475770789e-06,
      "loss": 0.1558,
      "num_input_tokens_seen": 48891448,
      "step": 30400
    },
    {
      "epoch": 1.3761968356006247,
      "eval_loss": 0.14403517544269562,
      "eval_runtime": 404.9214,
      "eval_samples_per_second": 96.984,
      "eval_steps_per_second": 24.247,
      "num_input_tokens_seen": 48891448,
      "step": 30400
    },
    {
      "epoch": 1.3764231874872677,
      "grad_norm": 0.40307581424713135,
      "learning_rate": 6.77040879348459e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 48899512,
      "step": 30405
    },
    {
      "epoch": 1.3766495393739107,
      "grad_norm": 0.4087376296520233,
      "learning_rate": 6.763691922430443e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 48907576,
      "step": 30410
    },
    {
      "epoch": 1.3768758912605537,
      "grad_norm": 0.5949267745018005,
      "learning_rate": 6.756977863644178e-06,
      "loss": 0.1107,
      "num_input_tokens_seen": 48915672,
      "step": 30415
    },
    {
      "epoch": 1.3771022431471966,
      "grad_norm": 0.49547910690307617,
      "learning_rate": 6.7502666181611804e-06,
      "loss": 0.1453,
      "num_input_tokens_seen": 48923736,
      "step": 30420
    },
    {
      "epoch": 1.3773285950338396,
      "grad_norm": 1.0371553897857666,
      "learning_rate": 6.743558187016405e-06,
      "loss": 0.1556,
      "num_input_tokens_seen": 48931576,
      "step": 30425
    },
    {
      "epoch": 1.3775549469204826,
      "grad_norm": 0.48403289914131165,
      "learning_rate": 6.7368525712443925e-06,
      "loss": 0.1659,
      "num_input_tokens_seen": 48940760,
      "step": 30430
    },
    {
      "epoch": 1.3777812988071256,
      "grad_norm": 0.4712201952934265,
      "learning_rate": 6.7301497718792155e-06,
      "loss": 0.1165,
      "num_input_tokens_seen": 48949464,
      "step": 30435
    },
    {
      "epoch": 1.3780076506937684,
      "grad_norm": 0.5137351155281067,
      "learning_rate": 6.723449789954544e-06,
      "loss": 0.1274,
      "num_input_tokens_seen": 48956760,
      "step": 30440
    },
    {
      "epoch": 1.3782340025804114,
      "grad_norm": 0.2928834855556488,
      "learning_rate": 6.716752626503586e-06,
      "loss": 0.1083,
      "num_input_tokens_seen": 48964472,
      "step": 30445
    },
    {
      "epoch": 1.3784603544670544,
      "grad_norm": 0.5114205479621887,
      "learning_rate": 6.710058282559131e-06,
      "loss": 0.1466,
      "num_input_tokens_seen": 48972120,
      "step": 30450
    },
    {
      "epoch": 1.3786867063536974,
      "grad_norm": 0.9218868613243103,
      "learning_rate": 6.703366759153545e-06,
      "loss": 0.1561,
      "num_input_tokens_seen": 48980056,
      "step": 30455
    },
    {
      "epoch": 1.3789130582403404,
      "grad_norm": 0.41436877846717834,
      "learning_rate": 6.6966780573187335e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 48988376,
      "step": 30460
    },
    {
      "epoch": 1.3791394101269834,
      "grad_norm": 0.35470399260520935,
      "learning_rate": 6.689992178086174e-06,
      "loss": 0.1081,
      "num_input_tokens_seen": 48996504,
      "step": 30465
    },
    {
      "epoch": 1.3793657620136264,
      "grad_norm": 0.6438407897949219,
      "learning_rate": 6.683309122486925e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 49005016,
      "step": 30470
    },
    {
      "epoch": 1.3795921139002694,
      "grad_norm": 0.6382662653923035,
      "learning_rate": 6.676628891551584e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 49014200,
      "step": 30475
    },
    {
      "epoch": 1.3798184657869124,
      "grad_norm": 0.7959979772567749,
      "learning_rate": 6.6699514863103385e-06,
      "loss": 0.161,
      "num_input_tokens_seen": 49022200,
      "step": 30480
    },
    {
      "epoch": 1.3800448176735554,
      "grad_norm": 0.48037436604499817,
      "learning_rate": 6.663276907792921e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 49030840,
      "step": 30485
    },
    {
      "epoch": 1.3802711695601984,
      "grad_norm": 0.7077735066413879,
      "learning_rate": 6.656605157028634e-06,
      "loss": 0.1209,
      "num_input_tokens_seen": 49038584,
      "step": 30490
    },
    {
      "epoch": 1.3804975214468413,
      "grad_norm": 0.289669930934906,
      "learning_rate": 6.649936235046358e-06,
      "loss": 0.1407,
      "num_input_tokens_seen": 49046392,
      "step": 30495
    },
    {
      "epoch": 1.3807238733334843,
      "grad_norm": 0.296915203332901,
      "learning_rate": 6.643270142874508e-06,
      "loss": 0.1138,
      "num_input_tokens_seen": 49054232,
      "step": 30500
    },
    {
      "epoch": 1.3809502252201273,
      "grad_norm": 0.6065211892127991,
      "learning_rate": 6.636606881541094e-06,
      "loss": 0.1218,
      "num_input_tokens_seen": 49061880,
      "step": 30505
    },
    {
      "epoch": 1.38117657710677,
      "grad_norm": 0.980369508266449,
      "learning_rate": 6.629946452073662e-06,
      "loss": 0.154,
      "num_input_tokens_seen": 49070072,
      "step": 30510
    },
    {
      "epoch": 1.381402928993413,
      "grad_norm": 0.4948841631412506,
      "learning_rate": 6.6232888554993375e-06,
      "loss": 0.1187,
      "num_input_tokens_seen": 49078136,
      "step": 30515
    },
    {
      "epoch": 1.381629280880056,
      "grad_norm": 0.38909804821014404,
      "learning_rate": 6.616634092844817e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 49086072,
      "step": 30520
    },
    {
      "epoch": 1.381855632766699,
      "grad_norm": 0.9111799597740173,
      "learning_rate": 6.609982165136331e-06,
      "loss": 0.1333,
      "num_input_tokens_seen": 49093816,
      "step": 30525
    },
    {
      "epoch": 1.382081984653342,
      "grad_norm": 0.6565498113632202,
      "learning_rate": 6.603333073399706e-06,
      "loss": 0.1632,
      "num_input_tokens_seen": 49101272,
      "step": 30530
    },
    {
      "epoch": 1.382308336539985,
      "grad_norm": 0.570449709892273,
      "learning_rate": 6.596686818660308e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 49109208,
      "step": 30535
    },
    {
      "epoch": 1.382534688426628,
      "grad_norm": 0.48102831840515137,
      "learning_rate": 6.590043401943066e-06,
      "loss": 0.122,
      "num_input_tokens_seen": 49117144,
      "step": 30540
    },
    {
      "epoch": 1.382761040313271,
      "grad_norm": 0.6982148289680481,
      "learning_rate": 6.583402824272494e-06,
      "loss": 0.1331,
      "num_input_tokens_seen": 49125688,
      "step": 30545
    },
    {
      "epoch": 1.382987392199914,
      "grad_norm": 0.3108547627925873,
      "learning_rate": 6.576765086672634e-06,
      "loss": 0.126,
      "num_input_tokens_seen": 49133976,
      "step": 30550
    },
    {
      "epoch": 1.3832137440865568,
      "grad_norm": 0.5333953499794006,
      "learning_rate": 6.57013019016712e-06,
      "loss": 0.1548,
      "num_input_tokens_seen": 49141624,
      "step": 30555
    },
    {
      "epoch": 1.3834400959731998,
      "grad_norm": 0.610637366771698,
      "learning_rate": 6.563498135779142e-06,
      "loss": 0.124,
      "num_input_tokens_seen": 49149624,
      "step": 30560
    },
    {
      "epoch": 1.3836664478598428,
      "grad_norm": 0.9001417756080627,
      "learning_rate": 6.556868924531431e-06,
      "loss": 0.1685,
      "num_input_tokens_seen": 49157976,
      "step": 30565
    },
    {
      "epoch": 1.3838927997464858,
      "grad_norm": 0.41970494389533997,
      "learning_rate": 6.550242557446304e-06,
      "loss": 0.1076,
      "num_input_tokens_seen": 49166744,
      "step": 30570
    },
    {
      "epoch": 1.3841191516331288,
      "grad_norm": 0.8032317757606506,
      "learning_rate": 6.543619035545634e-06,
      "loss": 0.159,
      "num_input_tokens_seen": 49174456,
      "step": 30575
    },
    {
      "epoch": 1.3843455035197718,
      "grad_norm": 1.104093074798584,
      "learning_rate": 6.53699835985084e-06,
      "loss": 0.1509,
      "num_input_tokens_seen": 49182392,
      "step": 30580
    },
    {
      "epoch": 1.3845718554064148,
      "grad_norm": 0.5130487084388733,
      "learning_rate": 6.530380531382927e-06,
      "loss": 0.1672,
      "num_input_tokens_seen": 49191608,
      "step": 30585
    },
    {
      "epoch": 1.3847982072930578,
      "grad_norm": 0.48881208896636963,
      "learning_rate": 6.523765551162433e-06,
      "loss": 0.1466,
      "num_input_tokens_seen": 49199160,
      "step": 30590
    },
    {
      "epoch": 1.3850245591797008,
      "grad_norm": 1.129298210144043,
      "learning_rate": 6.517153420209476e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 49207608,
      "step": 30595
    },
    {
      "epoch": 1.3852509110663438,
      "grad_norm": 1.6445703506469727,
      "learning_rate": 6.510544139543739e-06,
      "loss": 0.1246,
      "num_input_tokens_seen": 49216568,
      "step": 30600
    },
    {
      "epoch": 1.3852509110663438,
      "eval_loss": 0.14374695718288422,
      "eval_runtime": 404.4283,
      "eval_samples_per_second": 97.103,
      "eval_steps_per_second": 24.276,
      "num_input_tokens_seen": 49216568,
      "step": 30600
    },
    {
      "epoch": 1.3854772629529868,
      "grad_norm": 0.643379271030426,
      "learning_rate": 6.503937710184452e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 49224664,
      "step": 30605
    },
    {
      "epoch": 1.3857036148396298,
      "grad_norm": 0.9272206425666809,
      "learning_rate": 6.4973341331503954e-06,
      "loss": 0.1813,
      "num_input_tokens_seen": 49232632,
      "step": 30610
    },
    {
      "epoch": 1.3859299667262728,
      "grad_norm": 1.0302298069000244,
      "learning_rate": 6.490733409459942e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 49240824,
      "step": 30615
    },
    {
      "epoch": 1.3861563186129158,
      "grad_norm": 0.275982141494751,
      "learning_rate": 6.484135540130995e-06,
      "loss": 0.1298,
      "num_input_tokens_seen": 49248760,
      "step": 30620
    },
    {
      "epoch": 1.3863826704995585,
      "grad_norm": 0.9719597697257996,
      "learning_rate": 6.4775405261810364e-06,
      "loss": 0.1909,
      "num_input_tokens_seen": 49256856,
      "step": 30625
    },
    {
      "epoch": 1.3866090223862015,
      "grad_norm": 0.799709677696228,
      "learning_rate": 6.470948368627092e-06,
      "loss": 0.1426,
      "num_input_tokens_seen": 49265464,
      "step": 30630
    },
    {
      "epoch": 1.3868353742728445,
      "grad_norm": 0.8494701981544495,
      "learning_rate": 6.464359068485756e-06,
      "loss": 0.1178,
      "num_input_tokens_seen": 49273624,
      "step": 30635
    },
    {
      "epoch": 1.3870617261594875,
      "grad_norm": 0.3136011064052582,
      "learning_rate": 6.457772626773195e-06,
      "loss": 0.1337,
      "num_input_tokens_seen": 49282072,
      "step": 30640
    },
    {
      "epoch": 1.3872880780461305,
      "grad_norm": 1.018303632736206,
      "learning_rate": 6.451189044505104e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 49290040,
      "step": 30645
    },
    {
      "epoch": 1.3875144299327735,
      "grad_norm": 0.769810676574707,
      "learning_rate": 6.44460832269676e-06,
      "loss": 0.1669,
      "num_input_tokens_seen": 49298520,
      "step": 30650
    },
    {
      "epoch": 1.3877407818194165,
      "grad_norm": 0.6408466100692749,
      "learning_rate": 6.438030462363001e-06,
      "loss": 0.1419,
      "num_input_tokens_seen": 49306296,
      "step": 30655
    },
    {
      "epoch": 1.3879671337060595,
      "grad_norm": 0.428514689207077,
      "learning_rate": 6.431455464518205e-06,
      "loss": 0.123,
      "num_input_tokens_seen": 49314200,
      "step": 30660
    },
    {
      "epoch": 1.3881934855927023,
      "grad_norm": 0.3757469952106476,
      "learning_rate": 6.424883330176326e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 49322264,
      "step": 30665
    },
    {
      "epoch": 1.3884198374793453,
      "grad_norm": 1.3801976442337036,
      "learning_rate": 6.418314060350864e-06,
      "loss": 0.1589,
      "num_input_tokens_seen": 49330264,
      "step": 30670
    },
    {
      "epoch": 1.3886461893659883,
      "grad_norm": 0.37440118193626404,
      "learning_rate": 6.4117476560548895e-06,
      "loss": 0.142,
      "num_input_tokens_seen": 49338168,
      "step": 30675
    },
    {
      "epoch": 1.3888725412526313,
      "grad_norm": 0.3090144991874695,
      "learning_rate": 6.405184118301016e-06,
      "loss": 0.1424,
      "num_input_tokens_seen": 49345848,
      "step": 30680
    },
    {
      "epoch": 1.3890988931392743,
      "grad_norm": 0.4767855703830719,
      "learning_rate": 6.398623448101434e-06,
      "loss": 0.1035,
      "num_input_tokens_seen": 49353432,
      "step": 30685
    },
    {
      "epoch": 1.3893252450259173,
      "grad_norm": 0.6416718363761902,
      "learning_rate": 6.392065646467871e-06,
      "loss": 0.11,
      "num_input_tokens_seen": 49361560,
      "step": 30690
    },
    {
      "epoch": 1.3895515969125602,
      "grad_norm": 0.8756217956542969,
      "learning_rate": 6.385510714411632e-06,
      "loss": 0.1184,
      "num_input_tokens_seen": 49369560,
      "step": 30695
    },
    {
      "epoch": 1.3897779487992032,
      "grad_norm": 0.47978585958480835,
      "learning_rate": 6.378958652943559e-06,
      "loss": 0.1599,
      "num_input_tokens_seen": 49377656,
      "step": 30700
    },
    {
      "epoch": 1.3900043006858462,
      "grad_norm": 0.30211707949638367,
      "learning_rate": 6.3724094630740776e-06,
      "loss": 0.102,
      "num_input_tokens_seen": 49385240,
      "step": 30705
    },
    {
      "epoch": 1.3902306525724892,
      "grad_norm": 0.4542778432369232,
      "learning_rate": 6.365863145813136e-06,
      "loss": 0.1493,
      "num_input_tokens_seen": 49393304,
      "step": 30710
    },
    {
      "epoch": 1.3904570044591322,
      "grad_norm": 0.816311240196228,
      "learning_rate": 6.359319702170269e-06,
      "loss": 0.1195,
      "num_input_tokens_seen": 49401592,
      "step": 30715
    },
    {
      "epoch": 1.3906833563457752,
      "grad_norm": 0.43162015080451965,
      "learning_rate": 6.352779133154566e-06,
      "loss": 0.1084,
      "num_input_tokens_seen": 49409784,
      "step": 30720
    },
    {
      "epoch": 1.3909097082324182,
      "grad_norm": 0.506003737449646,
      "learning_rate": 6.346241439774648e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 49418008,
      "step": 30725
    },
    {
      "epoch": 1.3911360601190612,
      "grad_norm": 0.39490848779678345,
      "learning_rate": 6.339706623038716e-06,
      "loss": 0.1472,
      "num_input_tokens_seen": 49425304,
      "step": 30730
    },
    {
      "epoch": 1.391362412005704,
      "grad_norm": 0.5877684950828552,
      "learning_rate": 6.333174683954532e-06,
      "loss": 0.1284,
      "num_input_tokens_seen": 49432632,
      "step": 30735
    },
    {
      "epoch": 1.391588763892347,
      "grad_norm": 0.814213752746582,
      "learning_rate": 6.326645623529387e-06,
      "loss": 0.1268,
      "num_input_tokens_seen": 49440120,
      "step": 30740
    },
    {
      "epoch": 1.39181511577899,
      "grad_norm": 0.5526827573776245,
      "learning_rate": 6.320119442770156e-06,
      "loss": 0.1588,
      "num_input_tokens_seen": 49447896,
      "step": 30745
    },
    {
      "epoch": 1.392041467665633,
      "grad_norm": 0.6840607523918152,
      "learning_rate": 6.313596142683254e-06,
      "loss": 0.1533,
      "num_input_tokens_seen": 49456216,
      "step": 30750
    },
    {
      "epoch": 1.392267819552276,
      "grad_norm": 0.6191989779472351,
      "learning_rate": 6.307075724274647e-06,
      "loss": 0.1541,
      "num_input_tokens_seen": 49464088,
      "step": 30755
    },
    {
      "epoch": 1.392494171438919,
      "grad_norm": 1.392421841621399,
      "learning_rate": 6.300558188549882e-06,
      "loss": 0.1599,
      "num_input_tokens_seen": 49471960,
      "step": 30760
    },
    {
      "epoch": 1.392720523325562,
      "grad_norm": 0.4518950283527374,
      "learning_rate": 6.29404353651403e-06,
      "loss": 0.1808,
      "num_input_tokens_seen": 49479704,
      "step": 30765
    },
    {
      "epoch": 1.392946875212205,
      "grad_norm": 0.2244957834482193,
      "learning_rate": 6.287531769171737e-06,
      "loss": 0.104,
      "num_input_tokens_seen": 49487256,
      "step": 30770
    },
    {
      "epoch": 1.393173227098848,
      "grad_norm": 0.5683499574661255,
      "learning_rate": 6.2810228875272045e-06,
      "loss": 0.184,
      "num_input_tokens_seen": 49495544,
      "step": 30775
    },
    {
      "epoch": 1.3933995789854907,
      "grad_norm": 0.8818849325180054,
      "learning_rate": 6.274516892584179e-06,
      "loss": 0.1161,
      "num_input_tokens_seen": 49504280,
      "step": 30780
    },
    {
      "epoch": 1.3936259308721337,
      "grad_norm": 0.9894168376922607,
      "learning_rate": 6.268013785345969e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 49511896,
      "step": 30785
    },
    {
      "epoch": 1.3938522827587767,
      "grad_norm": 0.6657827496528625,
      "learning_rate": 6.26151356681543e-06,
      "loss": 0.1786,
      "num_input_tokens_seen": 49519416,
      "step": 30790
    },
    {
      "epoch": 1.3940786346454197,
      "grad_norm": 0.6724255681037903,
      "learning_rate": 6.255016237994981e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 49527608,
      "step": 30795
    },
    {
      "epoch": 1.3943049865320627,
      "grad_norm": 0.4994489550590515,
      "learning_rate": 6.248521799886603e-06,
      "loss": 0.1452,
      "num_input_tokens_seen": 49535352,
      "step": 30800
    },
    {
      "epoch": 1.3943049865320627,
      "eval_loss": 0.14399762451648712,
      "eval_runtime": 405.0325,
      "eval_samples_per_second": 96.958,
      "eval_steps_per_second": 24.24,
      "num_input_tokens_seen": 49535352,
      "step": 30800
    },
    {
      "epoch": 1.3945313384187057,
      "grad_norm": 0.7318084239959717,
      "learning_rate": 6.242030253491798e-06,
      "loss": 0.1792,
      "num_input_tokens_seen": 49543448,
      "step": 30805
    },
    {
      "epoch": 1.3947576903053487,
      "grad_norm": 0.2573508322238922,
      "learning_rate": 6.235541599811656e-06,
      "loss": 0.1066,
      "num_input_tokens_seen": 49552312,
      "step": 30810
    },
    {
      "epoch": 1.3949840421919917,
      "grad_norm": 0.4910697340965271,
      "learning_rate": 6.229055839846814e-06,
      "loss": 0.1409,
      "num_input_tokens_seen": 49560600,
      "step": 30815
    },
    {
      "epoch": 1.3952103940786347,
      "grad_norm": 0.6019947528839111,
      "learning_rate": 6.222572974597455e-06,
      "loss": 0.093,
      "num_input_tokens_seen": 49568696,
      "step": 30820
    },
    {
      "epoch": 1.3954367459652777,
      "grad_norm": 0.7334957718849182,
      "learning_rate": 6.216093005063306e-06,
      "loss": 0.1231,
      "num_input_tokens_seen": 49576888,
      "step": 30825
    },
    {
      "epoch": 1.3956630978519207,
      "grad_norm": 0.4797588884830475,
      "learning_rate": 6.209615932243678e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 49585112,
      "step": 30830
    },
    {
      "epoch": 1.3958894497385637,
      "grad_norm": 1.2374662160873413,
      "learning_rate": 6.203141757137399e-06,
      "loss": 0.1633,
      "num_input_tokens_seen": 49592888,
      "step": 30835
    },
    {
      "epoch": 1.3961158016252067,
      "grad_norm": 0.778872549533844,
      "learning_rate": 6.196670480742886e-06,
      "loss": 0.1085,
      "num_input_tokens_seen": 49600504,
      "step": 30840
    },
    {
      "epoch": 1.3963421535118496,
      "grad_norm": 0.23552896082401276,
      "learning_rate": 6.190202104058074e-06,
      "loss": 0.1006,
      "num_input_tokens_seen": 49608504,
      "step": 30845
    },
    {
      "epoch": 1.3965685053984924,
      "grad_norm": 1.0558574199676514,
      "learning_rate": 6.183736628080475e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 49616728,
      "step": 30850
    },
    {
      "epoch": 1.3967948572851354,
      "grad_norm": 0.574908971786499,
      "learning_rate": 6.177274053807155e-06,
      "loss": 0.168,
      "num_input_tokens_seen": 49624376,
      "step": 30855
    },
    {
      "epoch": 1.3970212091717784,
      "grad_norm": 0.5534267425537109,
      "learning_rate": 6.170814382234713e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 49632280,
      "step": 30860
    },
    {
      "epoch": 1.3972475610584214,
      "grad_norm": 0.352756142616272,
      "learning_rate": 6.16435761435932e-06,
      "loss": 0.1168,
      "num_input_tokens_seen": 49640536,
      "step": 30865
    },
    {
      "epoch": 1.3974739129450644,
      "grad_norm": 0.4328562617301941,
      "learning_rate": 6.157903751176681e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 49648856,
      "step": 30870
    },
    {
      "epoch": 1.3977002648317074,
      "grad_norm": 0.5755569338798523,
      "learning_rate": 6.151452793682066e-06,
      "loss": 0.115,
      "num_input_tokens_seen": 49656696,
      "step": 30875
    },
    {
      "epoch": 1.3979266167183504,
      "grad_norm": 0.8543965816497803,
      "learning_rate": 6.145004742870305e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 49664536,
      "step": 30880
    },
    {
      "epoch": 1.3981529686049934,
      "grad_norm": 0.5498054027557373,
      "learning_rate": 6.138559599735752e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 49672440,
      "step": 30885
    },
    {
      "epoch": 1.3983793204916364,
      "grad_norm": 0.932090699672699,
      "learning_rate": 6.132117365272344e-06,
      "loss": 0.1146,
      "num_input_tokens_seen": 49680536,
      "step": 30890
    },
    {
      "epoch": 1.3986056723782792,
      "grad_norm": 0.5603472590446472,
      "learning_rate": 6.125678040473545e-06,
      "loss": 0.132,
      "num_input_tokens_seen": 49689208,
      "step": 30895
    },
    {
      "epoch": 1.3988320242649221,
      "grad_norm": 0.5687379240989685,
      "learning_rate": 6.1192416263323755e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 49696728,
      "step": 30900
    },
    {
      "epoch": 1.3990583761515651,
      "grad_norm": 0.9439936876296997,
      "learning_rate": 6.112808123841424e-06,
      "loss": 0.1637,
      "num_input_tokens_seen": 49704888,
      "step": 30905
    },
    {
      "epoch": 1.3992847280382081,
      "grad_norm": 0.44331100583076477,
      "learning_rate": 6.106377533992805e-06,
      "loss": 0.1467,
      "num_input_tokens_seen": 49712792,
      "step": 30910
    },
    {
      "epoch": 1.3995110799248511,
      "grad_norm": 0.8030891418457031,
      "learning_rate": 6.099949857778204e-06,
      "loss": 0.141,
      "num_input_tokens_seen": 49720856,
      "step": 30915
    },
    {
      "epoch": 1.3997374318114941,
      "grad_norm": 0.36840546131134033,
      "learning_rate": 6.093525096188852e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 49729272,
      "step": 30920
    },
    {
      "epoch": 1.3999637836981371,
      "grad_norm": 0.22602826356887817,
      "learning_rate": 6.087103250215518e-06,
      "loss": 0.107,
      "num_input_tokens_seen": 49737400,
      "step": 30925
    },
    {
      "epoch": 1.4001901355847801,
      "grad_norm": 1.213016152381897,
      "learning_rate": 6.080684320848537e-06,
      "loss": 0.1548,
      "num_input_tokens_seen": 49746104,
      "step": 30930
    },
    {
      "epoch": 1.400416487471423,
      "grad_norm": 0.4497641921043396,
      "learning_rate": 6.074268309077794e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 49754168,
      "step": 30935
    },
    {
      "epoch": 1.400642839358066,
      "grad_norm": 0.7082121968269348,
      "learning_rate": 6.067855215892709e-06,
      "loss": 0.1481,
      "num_input_tokens_seen": 49762328,
      "step": 30940
    },
    {
      "epoch": 1.400869191244709,
      "grad_norm": 0.4722375273704529,
      "learning_rate": 6.061445042282271e-06,
      "loss": 0.1567,
      "num_input_tokens_seen": 49770520,
      "step": 30945
    },
    {
      "epoch": 1.401095543131352,
      "grad_norm": 0.38528767228126526,
      "learning_rate": 6.055037789234999e-06,
      "loss": 0.1245,
      "num_input_tokens_seen": 49778584,
      "step": 30950
    },
    {
      "epoch": 1.401321895017995,
      "grad_norm": 1.1757429838180542,
      "learning_rate": 6.048633457738975e-06,
      "loss": 0.1424,
      "num_input_tokens_seen": 49787000,
      "step": 30955
    },
    {
      "epoch": 1.401548246904638,
      "grad_norm": 0.6900447607040405,
      "learning_rate": 6.042232048781837e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 49794776,
      "step": 30960
    },
    {
      "epoch": 1.4017745987912809,
      "grad_norm": 0.9946152567863464,
      "learning_rate": 6.035833563350757e-06,
      "loss": 0.1586,
      "num_input_tokens_seen": 49802808,
      "step": 30965
    },
    {
      "epoch": 1.4020009506779239,
      "grad_norm": 1.1505359411239624,
      "learning_rate": 6.0294380024324525e-06,
      "loss": 0.1402,
      "num_input_tokens_seen": 49810488,
      "step": 30970
    },
    {
      "epoch": 1.4022273025645668,
      "grad_norm": 0.6504638195037842,
      "learning_rate": 6.023045367013213e-06,
      "loss": 0.1598,
      "num_input_tokens_seen": 49818392,
      "step": 30975
    },
    {
      "epoch": 1.4024536544512098,
      "grad_norm": 0.8459160923957825,
      "learning_rate": 6.016655658078851e-06,
      "loss": 0.147,
      "num_input_tokens_seen": 49826200,
      "step": 30980
    },
    {
      "epoch": 1.4026800063378528,
      "grad_norm": 0.49962711334228516,
      "learning_rate": 6.010268876614753e-06,
      "loss": 0.1096,
      "num_input_tokens_seen": 49834264,
      "step": 30985
    },
    {
      "epoch": 1.4029063582244958,
      "grad_norm": 0.8901089429855347,
      "learning_rate": 6.0038850236058266e-06,
      "loss": 0.1496,
      "num_input_tokens_seen": 49844216,
      "step": 30990
    },
    {
      "epoch": 1.4031327101111388,
      "grad_norm": 0.41344261169433594,
      "learning_rate": 5.997504100036549e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 49852568,
      "step": 30995
    },
    {
      "epoch": 1.4033590619977818,
      "grad_norm": 0.5413503050804138,
      "learning_rate": 5.991126106890949e-06,
      "loss": 0.1331,
      "num_input_tokens_seen": 49859864,
      "step": 31000
    },
    {
      "epoch": 1.4033590619977818,
      "eval_loss": 0.14372922480106354,
      "eval_runtime": 405.5359,
      "eval_samples_per_second": 96.837,
      "eval_steps_per_second": 24.21,
      "num_input_tokens_seen": 49859864,
      "step": 31000
    },
    {
      "epoch": 1.4035854138844246,
      "grad_norm": 0.5597949624061584,
      "learning_rate": 5.984751045152576e-06,
      "loss": 0.1552,
      "num_input_tokens_seen": 49867672,
      "step": 31005
    },
    {
      "epoch": 1.4038117657710676,
      "grad_norm": 0.6817839741706848,
      "learning_rate": 5.978378915804553e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 49875640,
      "step": 31010
    },
    {
      "epoch": 1.4040381176577106,
      "grad_norm": 0.4116970896720886,
      "learning_rate": 5.972009719829547e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 49884120,
      "step": 31015
    },
    {
      "epoch": 1.4042644695443536,
      "grad_norm": 0.928680419921875,
      "learning_rate": 5.965643458209755e-06,
      "loss": 0.1401,
      "num_input_tokens_seen": 49891928,
      "step": 31020
    },
    {
      "epoch": 1.4044908214309966,
      "grad_norm": 0.5704149603843689,
      "learning_rate": 5.95928013192695e-06,
      "loss": 0.1223,
      "num_input_tokens_seen": 49899992,
      "step": 31025
    },
    {
      "epoch": 1.4047171733176396,
      "grad_norm": 0.45182090997695923,
      "learning_rate": 5.952919741962423e-06,
      "loss": 0.1491,
      "num_input_tokens_seen": 49908024,
      "step": 31030
    },
    {
      "epoch": 1.4049435252042826,
      "grad_norm": 0.9474608302116394,
      "learning_rate": 5.946562289297042e-06,
      "loss": 0.146,
      "num_input_tokens_seen": 49916568,
      "step": 31035
    },
    {
      "epoch": 1.4051698770909256,
      "grad_norm": 1.0077569484710693,
      "learning_rate": 5.9402077749111855e-06,
      "loss": 0.1795,
      "num_input_tokens_seen": 49924664,
      "step": 31040
    },
    {
      "epoch": 1.4053962289775686,
      "grad_norm": 0.4506681561470032,
      "learning_rate": 5.933856199784821e-06,
      "loss": 0.1413,
      "num_input_tokens_seen": 49932600,
      "step": 31045
    },
    {
      "epoch": 1.4056225808642115,
      "grad_norm": 0.38792920112609863,
      "learning_rate": 5.927507564897419e-06,
      "loss": 0.1257,
      "num_input_tokens_seen": 49941496,
      "step": 31050
    },
    {
      "epoch": 1.4058489327508545,
      "grad_norm": 1.1044546365737915,
      "learning_rate": 5.9211618712280395e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 49949080,
      "step": 31055
    },
    {
      "epoch": 1.4060752846374975,
      "grad_norm": 0.8375057578086853,
      "learning_rate": 5.914819119755255e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 49956824,
      "step": 31060
    },
    {
      "epoch": 1.4063016365241405,
      "grad_norm": 0.6889464855194092,
      "learning_rate": 5.908479311457205e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 49965016,
      "step": 31065
    },
    {
      "epoch": 1.4065279884107835,
      "grad_norm": 0.6907184720039368,
      "learning_rate": 5.902142447311559e-06,
      "loss": 0.1834,
      "num_input_tokens_seen": 49973368,
      "step": 31070
    },
    {
      "epoch": 1.4067543402974263,
      "grad_norm": 0.7110697031021118,
      "learning_rate": 5.895808528295546e-06,
      "loss": 0.1343,
      "num_input_tokens_seen": 49981240,
      "step": 31075
    },
    {
      "epoch": 1.4069806921840693,
      "grad_norm": 0.5545710325241089,
      "learning_rate": 5.889477555385941e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 49989496,
      "step": 31080
    },
    {
      "epoch": 1.4072070440707123,
      "grad_norm": 0.945901095867157,
      "learning_rate": 5.883149529559051e-06,
      "loss": 0.1697,
      "num_input_tokens_seen": 49997816,
      "step": 31085
    },
    {
      "epoch": 1.4074333959573553,
      "grad_norm": 0.44818100333213806,
      "learning_rate": 5.876824451790738e-06,
      "loss": 0.1993,
      "num_input_tokens_seen": 50005656,
      "step": 31090
    },
    {
      "epoch": 1.4076597478439983,
      "grad_norm": 0.6634775400161743,
      "learning_rate": 5.87050232305642e-06,
      "loss": 0.1819,
      "num_input_tokens_seen": 50014136,
      "step": 31095
    },
    {
      "epoch": 1.4078860997306413,
      "grad_norm": 0.6066328883171082,
      "learning_rate": 5.864183144331034e-06,
      "loss": 0.1198,
      "num_input_tokens_seen": 50021880,
      "step": 31100
    },
    {
      "epoch": 1.4081124516172843,
      "grad_norm": 0.47799503803253174,
      "learning_rate": 5.857866916589089e-06,
      "loss": 0.1546,
      "num_input_tokens_seen": 50030296,
      "step": 31105
    },
    {
      "epoch": 1.4083388035039273,
      "grad_norm": 0.603267252445221,
      "learning_rate": 5.8515536408046216e-06,
      "loss": 0.1459,
      "num_input_tokens_seen": 50038008,
      "step": 31110
    },
    {
      "epoch": 1.4085651553905703,
      "grad_norm": 1.1777766942977905,
      "learning_rate": 5.845243317951208e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 50045944,
      "step": 31115
    },
    {
      "epoch": 1.408791507277213,
      "grad_norm": 0.4396110475063324,
      "learning_rate": 5.838935949001997e-06,
      "loss": 0.1277,
      "num_input_tokens_seen": 50054520,
      "step": 31120
    },
    {
      "epoch": 1.409017859163856,
      "grad_norm": 0.5234498977661133,
      "learning_rate": 5.8326315349296476e-06,
      "loss": 0.1339,
      "num_input_tokens_seen": 50063096,
      "step": 31125
    },
    {
      "epoch": 1.409244211050499,
      "grad_norm": 1.0809307098388672,
      "learning_rate": 5.826330076706396e-06,
      "loss": 0.1673,
      "num_input_tokens_seen": 50071448,
      "step": 31130
    },
    {
      "epoch": 1.409470562937142,
      "grad_norm": 0.5021415948867798,
      "learning_rate": 5.820031575303988e-06,
      "loss": 0.1647,
      "num_input_tokens_seen": 50079256,
      "step": 31135
    },
    {
      "epoch": 1.409696914823785,
      "grad_norm": 0.481486976146698,
      "learning_rate": 5.813736031693745e-06,
      "loss": 0.1614,
      "num_input_tokens_seen": 50087096,
      "step": 31140
    },
    {
      "epoch": 1.409923266710428,
      "grad_norm": 0.41274940967559814,
      "learning_rate": 5.807443446846522e-06,
      "loss": 0.1284,
      "num_input_tokens_seen": 50095224,
      "step": 31145
    },
    {
      "epoch": 1.410149618597071,
      "grad_norm": 0.4131726026535034,
      "learning_rate": 5.801153821732699e-06,
      "loss": 0.1434,
      "num_input_tokens_seen": 50103416,
      "step": 31150
    },
    {
      "epoch": 1.410375970483714,
      "grad_norm": 1.0452289581298828,
      "learning_rate": 5.794867157322229e-06,
      "loss": 0.1649,
      "num_input_tokens_seen": 50110808,
      "step": 31155
    },
    {
      "epoch": 1.410602322370357,
      "grad_norm": 0.5778034925460815,
      "learning_rate": 5.788583454584593e-06,
      "loss": 0.0962,
      "num_input_tokens_seen": 50119096,
      "step": 31160
    },
    {
      "epoch": 1.410828674257,
      "grad_norm": 1.1425420045852661,
      "learning_rate": 5.7823027144888075e-06,
      "loss": 0.1682,
      "num_input_tokens_seen": 50127512,
      "step": 31165
    },
    {
      "epoch": 1.411055026143643,
      "grad_norm": 0.7230789065361023,
      "learning_rate": 5.776024938003455e-06,
      "loss": 0.1435,
      "num_input_tokens_seen": 50135832,
      "step": 31170
    },
    {
      "epoch": 1.411281378030286,
      "grad_norm": 0.475661963224411,
      "learning_rate": 5.7697501260966345e-06,
      "loss": 0.1424,
      "num_input_tokens_seen": 50144120,
      "step": 31175
    },
    {
      "epoch": 1.411507729916929,
      "grad_norm": 0.9339954257011414,
      "learning_rate": 5.7634782797360145e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 50151928,
      "step": 31180
    },
    {
      "epoch": 1.411734081803572,
      "grad_norm": 0.7109025716781616,
      "learning_rate": 5.757209399888777e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 50159192,
      "step": 31185
    },
    {
      "epoch": 1.4119604336902147,
      "grad_norm": 0.8185903429985046,
      "learning_rate": 5.750943487521679e-06,
      "loss": 0.1242,
      "num_input_tokens_seen": 50166712,
      "step": 31190
    },
    {
      "epoch": 1.4121867855768577,
      "grad_norm": 1.229814887046814,
      "learning_rate": 5.744680543600986e-06,
      "loss": 0.1461,
      "num_input_tokens_seen": 50174616,
      "step": 31195
    },
    {
      "epoch": 1.4124131374635007,
      "grad_norm": 0.4274640679359436,
      "learning_rate": 5.738420569092537e-06,
      "loss": 0.1442,
      "num_input_tokens_seen": 50183160,
      "step": 31200
    },
    {
      "epoch": 1.4124131374635007,
      "eval_loss": 0.14382949471473694,
      "eval_runtime": 404.7005,
      "eval_samples_per_second": 97.037,
      "eval_steps_per_second": 24.26,
      "num_input_tokens_seen": 50183160,
      "step": 31200
    },
    {
      "epoch": 1.4126394893501437,
      "grad_norm": 1.1656266450881958,
      "learning_rate": 5.732163564961684e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 50191576,
      "step": 31205
    },
    {
      "epoch": 1.4128658412367867,
      "grad_norm": 1.3231528997421265,
      "learning_rate": 5.725909532173354e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 50199256,
      "step": 31210
    },
    {
      "epoch": 1.4130921931234297,
      "grad_norm": 0.9327594637870789,
      "learning_rate": 5.719658471691977e-06,
      "loss": 0.1528,
      "num_input_tokens_seen": 50206808,
      "step": 31215
    },
    {
      "epoch": 1.4133185450100727,
      "grad_norm": 0.42784592509269714,
      "learning_rate": 5.71341038448156e-06,
      "loss": 0.1581,
      "num_input_tokens_seen": 50214840,
      "step": 31220
    },
    {
      "epoch": 1.4135448968967157,
      "grad_norm": 0.4829484820365906,
      "learning_rate": 5.707165271505635e-06,
      "loss": 0.1519,
      "num_input_tokens_seen": 50223192,
      "step": 31225
    },
    {
      "epoch": 1.4137712487833587,
      "grad_norm": 0.423143208026886,
      "learning_rate": 5.700923133727271e-06,
      "loss": 0.1482,
      "num_input_tokens_seen": 50230840,
      "step": 31230
    },
    {
      "epoch": 1.4139976006700015,
      "grad_norm": 0.9897170066833496,
      "learning_rate": 5.694683972109083e-06,
      "loss": 0.1577,
      "num_input_tokens_seen": 50238744,
      "step": 31235
    },
    {
      "epoch": 1.4142239525566445,
      "grad_norm": 0.3683108389377594,
      "learning_rate": 5.688447787613241e-06,
      "loss": 0.102,
      "num_input_tokens_seen": 50247000,
      "step": 31240
    },
    {
      "epoch": 1.4144503044432875,
      "grad_norm": 1.0745306015014648,
      "learning_rate": 5.6822145812014285e-06,
      "loss": 0.1684,
      "num_input_tokens_seen": 50255032,
      "step": 31245
    },
    {
      "epoch": 1.4146766563299304,
      "grad_norm": 0.6708888411521912,
      "learning_rate": 5.675984353834896e-06,
      "loss": 0.1487,
      "num_input_tokens_seen": 50262968,
      "step": 31250
    },
    {
      "epoch": 1.4149030082165734,
      "grad_norm": 0.43977630138397217,
      "learning_rate": 5.66975710647441e-06,
      "loss": 0.1204,
      "num_input_tokens_seen": 50270776,
      "step": 31255
    },
    {
      "epoch": 1.4151293601032164,
      "grad_norm": 0.8768766522407532,
      "learning_rate": 5.663532840080304e-06,
      "loss": 0.1267,
      "num_input_tokens_seen": 50279288,
      "step": 31260
    },
    {
      "epoch": 1.4153557119898594,
      "grad_norm": 0.6857749223709106,
      "learning_rate": 5.6573115556124325e-06,
      "loss": 0.1813,
      "num_input_tokens_seen": 50287032,
      "step": 31265
    },
    {
      "epoch": 1.4155820638765024,
      "grad_norm": 0.32195794582366943,
      "learning_rate": 5.651093254030185e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 50294936,
      "step": 31270
    },
    {
      "epoch": 1.4158084157631454,
      "grad_norm": 1.0630632638931274,
      "learning_rate": 5.644877936292514e-06,
      "loss": 0.1482,
      "num_input_tokens_seen": 50302968,
      "step": 31275
    },
    {
      "epoch": 1.4160347676497884,
      "grad_norm": 0.7014521360397339,
      "learning_rate": 5.638665603357901e-06,
      "loss": 0.1288,
      "num_input_tokens_seen": 50310968,
      "step": 31280
    },
    {
      "epoch": 1.4162611195364314,
      "grad_norm": 0.5945624709129333,
      "learning_rate": 5.632456256184357e-06,
      "loss": 0.1301,
      "num_input_tokens_seen": 50319128,
      "step": 31285
    },
    {
      "epoch": 1.4164874714230744,
      "grad_norm": 0.8677668571472168,
      "learning_rate": 5.626249895729452e-06,
      "loss": 0.1754,
      "num_input_tokens_seen": 50327096,
      "step": 31290
    },
    {
      "epoch": 1.4167138233097174,
      "grad_norm": 0.8659914135932922,
      "learning_rate": 5.620046522950273e-06,
      "loss": 0.1173,
      "num_input_tokens_seen": 50335576,
      "step": 31295
    },
    {
      "epoch": 1.4169401751963604,
      "grad_norm": 0.3515828549861908,
      "learning_rate": 5.613846138803464e-06,
      "loss": 0.1725,
      "num_input_tokens_seen": 50343512,
      "step": 31300
    },
    {
      "epoch": 1.4171665270830032,
      "grad_norm": 1.1590920686721802,
      "learning_rate": 5.607648744245206e-06,
      "loss": 0.1785,
      "num_input_tokens_seen": 50351640,
      "step": 31305
    },
    {
      "epoch": 1.4173928789696462,
      "grad_norm": 0.540358304977417,
      "learning_rate": 5.601454340231207e-06,
      "loss": 0.1566,
      "num_input_tokens_seen": 50360024,
      "step": 31310
    },
    {
      "epoch": 1.4176192308562892,
      "grad_norm": 0.4048596918582916,
      "learning_rate": 5.595262927716724e-06,
      "loss": 0.1493,
      "num_input_tokens_seen": 50367832,
      "step": 31315
    },
    {
      "epoch": 1.4178455827429322,
      "grad_norm": 0.9975108504295349,
      "learning_rate": 5.589074507656561e-06,
      "loss": 0.1591,
      "num_input_tokens_seen": 50375704,
      "step": 31320
    },
    {
      "epoch": 1.4180719346295751,
      "grad_norm": 0.4099293649196625,
      "learning_rate": 5.582889081005044e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 50384664,
      "step": 31325
    },
    {
      "epoch": 1.4182982865162181,
      "grad_norm": 0.4181903004646301,
      "learning_rate": 5.5767066487160316e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 50393240,
      "step": 31330
    },
    {
      "epoch": 1.4185246384028611,
      "grad_norm": 0.32744133472442627,
      "learning_rate": 5.570527211742949e-06,
      "loss": 0.1459,
      "num_input_tokens_seen": 50400984,
      "step": 31335
    },
    {
      "epoch": 1.4187509902895041,
      "grad_norm": 0.755579948425293,
      "learning_rate": 5.564350771038731e-06,
      "loss": 0.1158,
      "num_input_tokens_seen": 50409464,
      "step": 31340
    },
    {
      "epoch": 1.418977342176147,
      "grad_norm": 0.6224987506866455,
      "learning_rate": 5.558177327555875e-06,
      "loss": 0.1541,
      "num_input_tokens_seen": 50417304,
      "step": 31345
    },
    {
      "epoch": 1.41920369406279,
      "grad_norm": 0.30982494354248047,
      "learning_rate": 5.552006882246388e-06,
      "loss": 0.1169,
      "num_input_tokens_seen": 50425112,
      "step": 31350
    },
    {
      "epoch": 1.419430045949433,
      "grad_norm": 1.0749753713607788,
      "learning_rate": 5.545839436061839e-06,
      "loss": 0.1375,
      "num_input_tokens_seen": 50432792,
      "step": 31355
    },
    {
      "epoch": 1.419656397836076,
      "grad_norm": 0.6133634448051453,
      "learning_rate": 5.539674989953331e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 50440664,
      "step": 31360
    },
    {
      "epoch": 1.4198827497227189,
      "grad_norm": 0.9035732746124268,
      "learning_rate": 5.533513544871488e-06,
      "loss": 0.1422,
      "num_input_tokens_seen": 50449368,
      "step": 31365
    },
    {
      "epoch": 1.4201091016093619,
      "grad_norm": 0.9868890047073364,
      "learning_rate": 5.527355101766493e-06,
      "loss": 0.1314,
      "num_input_tokens_seen": 50457336,
      "step": 31370
    },
    {
      "epoch": 1.4203354534960049,
      "grad_norm": 0.5401564240455627,
      "learning_rate": 5.521199661588044e-06,
      "loss": 0.1208,
      "num_input_tokens_seen": 50465208,
      "step": 31375
    },
    {
      "epoch": 1.4205618053826479,
      "grad_norm": 0.5892345309257507,
      "learning_rate": 5.5150472252853944e-06,
      "loss": 0.1445,
      "num_input_tokens_seen": 50473176,
      "step": 31380
    },
    {
      "epoch": 1.4207881572692909,
      "grad_norm": 0.5610971450805664,
      "learning_rate": 5.50889779380733e-06,
      "loss": 0.1445,
      "num_input_tokens_seen": 50481240,
      "step": 31385
    },
    {
      "epoch": 1.4210145091559339,
      "grad_norm": 1.238152265548706,
      "learning_rate": 5.5027513681021605e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 50489144,
      "step": 31390
    },
    {
      "epoch": 1.4212408610425769,
      "grad_norm": 0.8913828134536743,
      "learning_rate": 5.4966079491177545e-06,
      "loss": 0.1537,
      "num_input_tokens_seen": 50497240,
      "step": 31395
    },
    {
      "epoch": 1.4214672129292198,
      "grad_norm": 0.38974514603614807,
      "learning_rate": 5.490467537801491e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 50505176,
      "step": 31400
    },
    {
      "epoch": 1.4214672129292198,
      "eval_loss": 0.1439216583967209,
      "eval_runtime": 404.832,
      "eval_samples_per_second": 97.006,
      "eval_steps_per_second": 24.252,
      "num_input_tokens_seen": 50505176,
      "step": 31400
    },
    {
      "epoch": 1.4216935648158628,
      "grad_norm": 1.5607274770736694,
      "learning_rate": 5.484330135100313e-06,
      "loss": 0.1576,
      "num_input_tokens_seen": 50513208,
      "step": 31405
    },
    {
      "epoch": 1.4219199167025058,
      "grad_norm": 0.7693782448768616,
      "learning_rate": 5.4781957419606785e-06,
      "loss": 0.1785,
      "num_input_tokens_seen": 50521400,
      "step": 31410
    },
    {
      "epoch": 1.4221462685891486,
      "grad_norm": 0.2164154052734375,
      "learning_rate": 5.472064359328577e-06,
      "loss": 0.1244,
      "num_input_tokens_seen": 50528952,
      "step": 31415
    },
    {
      "epoch": 1.4223726204757916,
      "grad_norm": 0.3491148352622986,
      "learning_rate": 5.4659359881495565e-06,
      "loss": 0.1235,
      "num_input_tokens_seen": 50536664,
      "step": 31420
    },
    {
      "epoch": 1.4225989723624346,
      "grad_norm": 0.6524737477302551,
      "learning_rate": 5.4598106293686916e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 50545112,
      "step": 31425
    },
    {
      "epoch": 1.4228253242490776,
      "grad_norm": 1.1377660036087036,
      "learning_rate": 5.45368828393058e-06,
      "loss": 0.213,
      "num_input_tokens_seen": 50553464,
      "step": 31430
    },
    {
      "epoch": 1.4230516761357206,
      "grad_norm": 1.1186786890029907,
      "learning_rate": 5.44756895277937e-06,
      "loss": 0.1773,
      "num_input_tokens_seen": 50561208,
      "step": 31435
    },
    {
      "epoch": 1.4232780280223636,
      "grad_norm": 1.0059431791305542,
      "learning_rate": 5.441452636858746e-06,
      "loss": 0.122,
      "num_input_tokens_seen": 50569208,
      "step": 31440
    },
    {
      "epoch": 1.4235043799090066,
      "grad_norm": 0.4075550436973572,
      "learning_rate": 5.435339337111905e-06,
      "loss": 0.126,
      "num_input_tokens_seen": 50576856,
      "step": 31445
    },
    {
      "epoch": 1.4237307317956496,
      "grad_norm": 0.4908052682876587,
      "learning_rate": 5.42922905448161e-06,
      "loss": 0.1671,
      "num_input_tokens_seen": 50585080,
      "step": 31450
    },
    {
      "epoch": 1.4239570836822926,
      "grad_norm": 1.0762284994125366,
      "learning_rate": 5.423121789910129e-06,
      "loss": 0.1507,
      "num_input_tokens_seen": 50592952,
      "step": 31455
    },
    {
      "epoch": 1.4241834355689353,
      "grad_norm": 1.331085443496704,
      "learning_rate": 5.417017544339287e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 50601752,
      "step": 31460
    },
    {
      "epoch": 1.4244097874555783,
      "grad_norm": 0.552642822265625,
      "learning_rate": 5.410916318710443e-06,
      "loss": 0.1276,
      "num_input_tokens_seen": 50609528,
      "step": 31465
    },
    {
      "epoch": 1.4246361393422213,
      "grad_norm": 0.34010860323905945,
      "learning_rate": 5.404818113964466e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 50618008,
      "step": 31470
    },
    {
      "epoch": 1.4248624912288643,
      "grad_norm": 0.2648994028568268,
      "learning_rate": 5.398722931041792e-06,
      "loss": 0.13,
      "num_input_tokens_seen": 50626456,
      "step": 31475
    },
    {
      "epoch": 1.4250888431155073,
      "grad_norm": 0.7235009670257568,
      "learning_rate": 5.392630770882367e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 50633912,
      "step": 31480
    },
    {
      "epoch": 1.4253151950021503,
      "grad_norm": 0.6739055514335632,
      "learning_rate": 5.3865416344256705e-06,
      "loss": 0.1448,
      "num_input_tokens_seen": 50641912,
      "step": 31485
    },
    {
      "epoch": 1.4255415468887933,
      "grad_norm": 0.7517442107200623,
      "learning_rate": 5.380455522610742e-06,
      "loss": 0.1811,
      "num_input_tokens_seen": 50649752,
      "step": 31490
    },
    {
      "epoch": 1.4257678987754363,
      "grad_norm": 1.0968685150146484,
      "learning_rate": 5.374372436376116e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 50657880,
      "step": 31495
    },
    {
      "epoch": 1.4259942506620793,
      "grad_norm": 1.0841519832611084,
      "learning_rate": 5.368292376659895e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 50665976,
      "step": 31500
    },
    {
      "epoch": 1.4262206025487223,
      "grad_norm": 0.5373808145523071,
      "learning_rate": 5.362215344399701e-06,
      "loss": 0.1343,
      "num_input_tokens_seen": 50674136,
      "step": 31505
    },
    {
      "epoch": 1.4264469544353653,
      "grad_norm": 0.3270271420478821,
      "learning_rate": 5.356141340532678e-06,
      "loss": 0.1409,
      "num_input_tokens_seen": 50682232,
      "step": 31510
    },
    {
      "epoch": 1.4266733063220083,
      "grad_norm": 0.30339521169662476,
      "learning_rate": 5.350070365995522e-06,
      "loss": 0.1444,
      "num_input_tokens_seen": 50690648,
      "step": 31515
    },
    {
      "epoch": 1.4268996582086513,
      "grad_norm": 0.9617205858230591,
      "learning_rate": 5.344002421724459e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 50698776,
      "step": 31520
    },
    {
      "epoch": 1.4271260100952943,
      "grad_norm": 0.4669232964515686,
      "learning_rate": 5.337937508655228e-06,
      "loss": 0.1626,
      "num_input_tokens_seen": 50706648,
      "step": 31525
    },
    {
      "epoch": 1.427352361981937,
      "grad_norm": 1.2335879802703857,
      "learning_rate": 5.331875627723126e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 50714424,
      "step": 31530
    },
    {
      "epoch": 1.42757871386858,
      "grad_norm": 1.3516091108322144,
      "learning_rate": 5.325816779862963e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 50722424,
      "step": 31535
    },
    {
      "epoch": 1.427805065755223,
      "grad_norm": 1.496442437171936,
      "learning_rate": 5.319760966009102e-06,
      "loss": 0.1467,
      "num_input_tokens_seen": 50730232,
      "step": 31540
    },
    {
      "epoch": 1.428031417641866,
      "grad_norm": 0.6131072044372559,
      "learning_rate": 5.3137081870954096e-06,
      "loss": 0.1298,
      "num_input_tokens_seen": 50737784,
      "step": 31545
    },
    {
      "epoch": 1.428257769528509,
      "grad_norm": 0.3529927730560303,
      "learning_rate": 5.307658444055313e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 50745656,
      "step": 31550
    },
    {
      "epoch": 1.428484121415152,
      "grad_norm": 0.5291609764099121,
      "learning_rate": 5.301611737821749e-06,
      "loss": 0.1236,
      "num_input_tokens_seen": 50753368,
      "step": 31555
    },
    {
      "epoch": 1.428710473301795,
      "grad_norm": 1.3023850917816162,
      "learning_rate": 5.295568069327206e-06,
      "loss": 0.1909,
      "num_input_tokens_seen": 50761368,
      "step": 31560
    },
    {
      "epoch": 1.428936825188438,
      "grad_norm": 0.7390946745872498,
      "learning_rate": 5.289527439503683e-06,
      "loss": 0.1115,
      "num_input_tokens_seen": 50769176,
      "step": 31565
    },
    {
      "epoch": 1.429163177075081,
      "grad_norm": 0.6018723845481873,
      "learning_rate": 5.28348984928273e-06,
      "loss": 0.1555,
      "num_input_tokens_seen": 50777112,
      "step": 31570
    },
    {
      "epoch": 1.4293895289617238,
      "grad_norm": 0.3527194857597351,
      "learning_rate": 5.27745529959541e-06,
      "loss": 0.0989,
      "num_input_tokens_seen": 50784664,
      "step": 31575
    },
    {
      "epoch": 1.4296158808483668,
      "grad_norm": 0.5399025082588196,
      "learning_rate": 5.271423791372335e-06,
      "loss": 0.116,
      "num_input_tokens_seen": 50792216,
      "step": 31580
    },
    {
      "epoch": 1.4298422327350098,
      "grad_norm": 0.8851398825645447,
      "learning_rate": 5.26539532554364e-06,
      "loss": 0.1264,
      "num_input_tokens_seen": 50800056,
      "step": 31585
    },
    {
      "epoch": 1.4300685846216528,
      "grad_norm": 0.3961818814277649,
      "learning_rate": 5.25936990303898e-06,
      "loss": 0.1172,
      "num_input_tokens_seen": 50807768,
      "step": 31590
    },
    {
      "epoch": 1.4302949365082958,
      "grad_norm": 0.40348052978515625,
      "learning_rate": 5.253347524787555e-06,
      "loss": 0.142,
      "num_input_tokens_seen": 50815704,
      "step": 31595
    },
    {
      "epoch": 1.4305212883949388,
      "grad_norm": 0.637630045413971,
      "learning_rate": 5.2473281917181035e-06,
      "loss": 0.1283,
      "num_input_tokens_seen": 50823608,
      "step": 31600
    },
    {
      "epoch": 1.4305212883949388,
      "eval_loss": 0.1441119760274887,
      "eval_runtime": 405.9226,
      "eval_samples_per_second": 96.745,
      "eval_steps_per_second": 24.187,
      "num_input_tokens_seen": 50823608,
      "step": 31600
    },
    {
      "epoch": 1.4307476402815817,
      "grad_norm": 0.6894921660423279,
      "learning_rate": 5.241311904758864e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 50831864,
      "step": 31605
    },
    {
      "epoch": 1.4309739921682247,
      "grad_norm": 1.070778489112854,
      "learning_rate": 5.23529866483764e-06,
      "loss": 0.1695,
      "num_input_tokens_seen": 50839416,
      "step": 31610
    },
    {
      "epoch": 1.4312003440548677,
      "grad_norm": 0.7987618446350098,
      "learning_rate": 5.229288472881732e-06,
      "loss": 0.1673,
      "num_input_tokens_seen": 50847992,
      "step": 31615
    },
    {
      "epoch": 1.4314266959415107,
      "grad_norm": 0.652528703212738,
      "learning_rate": 5.2232813298180025e-06,
      "loss": 0.1646,
      "num_input_tokens_seen": 50856024,
      "step": 31620
    },
    {
      "epoch": 1.4316530478281537,
      "grad_norm": 0.41726455092430115,
      "learning_rate": 5.217277236572824e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 50864152,
      "step": 31625
    },
    {
      "epoch": 1.4318793997147967,
      "grad_norm": 0.34039315581321716,
      "learning_rate": 5.211276194072093e-06,
      "loss": 0.2028,
      "num_input_tokens_seen": 50871992,
      "step": 31630
    },
    {
      "epoch": 1.4321057516014397,
      "grad_norm": 0.7886508703231812,
      "learning_rate": 5.205278203241254e-06,
      "loss": 0.1442,
      "num_input_tokens_seen": 50880408,
      "step": 31635
    },
    {
      "epoch": 1.4323321034880827,
      "grad_norm": 0.5312203168869019,
      "learning_rate": 5.199283265005278e-06,
      "loss": 0.1537,
      "num_input_tokens_seen": 50888344,
      "step": 31640
    },
    {
      "epoch": 1.4325584553747255,
      "grad_norm": 0.2605816125869751,
      "learning_rate": 5.193291380288648e-06,
      "loss": 0.0922,
      "num_input_tokens_seen": 50896856,
      "step": 31645
    },
    {
      "epoch": 1.4327848072613685,
      "grad_norm": 0.4617220163345337,
      "learning_rate": 5.1873025500153995e-06,
      "loss": 0.0932,
      "num_input_tokens_seen": 50904728,
      "step": 31650
    },
    {
      "epoch": 1.4330111591480115,
      "grad_norm": 0.5961469411849976,
      "learning_rate": 5.181316775109071e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 50912696,
      "step": 31655
    },
    {
      "epoch": 1.4332375110346545,
      "grad_norm": 0.5069565773010254,
      "learning_rate": 5.1753340564927564e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 50920152,
      "step": 31660
    },
    {
      "epoch": 1.4334638629212975,
      "grad_norm": 0.5300441384315491,
      "learning_rate": 5.169354395089068e-06,
      "loss": 0.1506,
      "num_input_tokens_seen": 50928408,
      "step": 31665
    },
    {
      "epoch": 1.4336902148079405,
      "grad_norm": 0.5993013978004456,
      "learning_rate": 5.1633777918201346e-06,
      "loss": 0.1376,
      "num_input_tokens_seen": 50936664,
      "step": 31670
    },
    {
      "epoch": 1.4339165666945834,
      "grad_norm": 0.411053329706192,
      "learning_rate": 5.157404247607625e-06,
      "loss": 0.1101,
      "num_input_tokens_seen": 50944504,
      "step": 31675
    },
    {
      "epoch": 1.4341429185812264,
      "grad_norm": 0.7293581962585449,
      "learning_rate": 5.1514337633727454e-06,
      "loss": 0.1463,
      "num_input_tokens_seen": 50953016,
      "step": 31680
    },
    {
      "epoch": 1.4343692704678692,
      "grad_norm": 0.6822235584259033,
      "learning_rate": 5.145466340036206e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 50960568,
      "step": 31685
    },
    {
      "epoch": 1.4345956223545122,
      "grad_norm": 0.35141897201538086,
      "learning_rate": 5.139501978518274e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 50968472,
      "step": 31690
    },
    {
      "epoch": 1.4348219742411552,
      "grad_norm": 0.5772151947021484,
      "learning_rate": 5.133540679738716e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 50976696,
      "step": 31695
    },
    {
      "epoch": 1.4350483261277982,
      "grad_norm": 1.038545846939087,
      "learning_rate": 5.127582444616838e-06,
      "loss": 0.1832,
      "num_input_tokens_seen": 50984184,
      "step": 31700
    },
    {
      "epoch": 1.4352746780144412,
      "grad_norm": 0.33497920632362366,
      "learning_rate": 5.121627274071486e-06,
      "loss": 0.179,
      "num_input_tokens_seen": 50992312,
      "step": 31705
    },
    {
      "epoch": 1.4355010299010842,
      "grad_norm": 0.8148412704467773,
      "learning_rate": 5.115675169021009e-06,
      "loss": 0.1206,
      "num_input_tokens_seen": 51000344,
      "step": 31710
    },
    {
      "epoch": 1.4357273817877272,
      "grad_norm": 0.5975478887557983,
      "learning_rate": 5.1097261303832994e-06,
      "loss": 0.1354,
      "num_input_tokens_seen": 51008152,
      "step": 31715
    },
    {
      "epoch": 1.4359537336743702,
      "grad_norm": 0.5084453225135803,
      "learning_rate": 5.103780159075788e-06,
      "loss": 0.1293,
      "num_input_tokens_seen": 51016408,
      "step": 31720
    },
    {
      "epoch": 1.4361800855610132,
      "grad_norm": 1.0553282499313354,
      "learning_rate": 5.0978372560154e-06,
      "loss": 0.1694,
      "num_input_tokens_seen": 51023960,
      "step": 31725
    },
    {
      "epoch": 1.4364064374476562,
      "grad_norm": 0.890432596206665,
      "learning_rate": 5.091897422118619e-06,
      "loss": 0.1163,
      "num_input_tokens_seen": 51032344,
      "step": 31730
    },
    {
      "epoch": 1.4366327893342992,
      "grad_norm": 0.7136051058769226,
      "learning_rate": 5.0859606583014305e-06,
      "loss": 0.1551,
      "num_input_tokens_seen": 51040376,
      "step": 31735
    },
    {
      "epoch": 1.4368591412209422,
      "grad_norm": 0.43691110610961914,
      "learning_rate": 5.080026965479365e-06,
      "loss": 0.1677,
      "num_input_tokens_seen": 51048312,
      "step": 31740
    },
    {
      "epoch": 1.4370854931075852,
      "grad_norm": 0.3845808207988739,
      "learning_rate": 5.074096344567475e-06,
      "loss": 0.1349,
      "num_input_tokens_seen": 51056184,
      "step": 31745
    },
    {
      "epoch": 1.4373118449942281,
      "grad_norm": 0.6398140788078308,
      "learning_rate": 5.0681687964803294e-06,
      "loss": 0.1972,
      "num_input_tokens_seen": 51064568,
      "step": 31750
    },
    {
      "epoch": 1.437538196880871,
      "grad_norm": 0.4117770791053772,
      "learning_rate": 5.06224432213204e-06,
      "loss": 0.1568,
      "num_input_tokens_seen": 51072984,
      "step": 31755
    },
    {
      "epoch": 1.437764548767514,
      "grad_norm": 0.8917523622512817,
      "learning_rate": 5.056322922436224e-06,
      "loss": 0.1253,
      "num_input_tokens_seen": 51080536,
      "step": 31760
    },
    {
      "epoch": 1.437990900654157,
      "grad_norm": 0.8372820615768433,
      "learning_rate": 5.0504045983060465e-06,
      "loss": 0.1438,
      "num_input_tokens_seen": 51088120,
      "step": 31765
    },
    {
      "epoch": 1.4382172525408,
      "grad_norm": 0.34468570351600647,
      "learning_rate": 5.044489350654183e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 51095768,
      "step": 31770
    },
    {
      "epoch": 1.438443604427443,
      "grad_norm": 0.801547110080719,
      "learning_rate": 5.038577180392831e-06,
      "loss": 0.1205,
      "num_input_tokens_seen": 51103064,
      "step": 31775
    },
    {
      "epoch": 1.438669956314086,
      "grad_norm": 0.6691551208496094,
      "learning_rate": 5.032668088433729e-06,
      "loss": 0.0914,
      "num_input_tokens_seen": 51110520,
      "step": 31780
    },
    {
      "epoch": 1.438896308200729,
      "grad_norm": 1.528236985206604,
      "learning_rate": 5.02676207568814e-06,
      "loss": 0.1743,
      "num_input_tokens_seen": 51118104,
      "step": 31785
    },
    {
      "epoch": 1.4391226600873719,
      "grad_norm": 0.6700114607810974,
      "learning_rate": 5.02085914306683e-06,
      "loss": 0.1634,
      "num_input_tokens_seen": 51125976,
      "step": 31790
    },
    {
      "epoch": 1.4393490119740149,
      "grad_norm": 0.2741040289402008,
      "learning_rate": 5.014959291480123e-06,
      "loss": 0.1133,
      "num_input_tokens_seen": 51133976,
      "step": 31795
    },
    {
      "epoch": 1.4395753638606577,
      "grad_norm": 0.7094343304634094,
      "learning_rate": 5.009062521837835e-06,
      "loss": 0.1138,
      "num_input_tokens_seen": 51142072,
      "step": 31800
    },
    {
      "epoch": 1.4395753638606577,
      "eval_loss": 0.14417912065982819,
      "eval_runtime": 404.6055,
      "eval_samples_per_second": 97.06,
      "eval_steps_per_second": 24.266,
      "num_input_tokens_seen": 51142072,
      "step": 31800
    },
    {
      "epoch": 1.4398017157473006,
      "grad_norm": 0.3285919427871704,
      "learning_rate": 5.003168835049324e-06,
      "loss": 0.1441,
      "num_input_tokens_seen": 51150584,
      "step": 31805
    },
    {
      "epoch": 1.4400280676339436,
      "grad_norm": 0.31568580865859985,
      "learning_rate": 4.997278232023483e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 51158392,
      "step": 31810
    },
    {
      "epoch": 1.4402544195205866,
      "grad_norm": 0.5829453468322754,
      "learning_rate": 4.9913907136687036e-06,
      "loss": 0.1308,
      "num_input_tokens_seen": 51166424,
      "step": 31815
    },
    {
      "epoch": 1.4404807714072296,
      "grad_norm": 0.6400544047355652,
      "learning_rate": 4.985506280892918e-06,
      "loss": 0.1302,
      "num_input_tokens_seen": 51175064,
      "step": 31820
    },
    {
      "epoch": 1.4407071232938726,
      "grad_norm": 0.5309350490570068,
      "learning_rate": 4.979624934603589e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 51183192,
      "step": 31825
    },
    {
      "epoch": 1.4409334751805156,
      "grad_norm": 0.7893242239952087,
      "learning_rate": 4.97374667570768e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 51191640,
      "step": 31830
    },
    {
      "epoch": 1.4411598270671586,
      "grad_norm": 1.0475120544433594,
      "learning_rate": 4.967871505111704e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 51199512,
      "step": 31835
    },
    {
      "epoch": 1.4413861789538016,
      "grad_norm": 0.3616182208061218,
      "learning_rate": 4.961999423721686e-06,
      "loss": 0.1416,
      "num_input_tokens_seen": 51207416,
      "step": 31840
    },
    {
      "epoch": 1.4416125308404446,
      "grad_norm": 0.4207071363925934,
      "learning_rate": 4.956130432443159e-06,
      "loss": 0.1489,
      "num_input_tokens_seen": 51215288,
      "step": 31845
    },
    {
      "epoch": 1.4418388827270876,
      "grad_norm": 1.0900357961654663,
      "learning_rate": 4.950264532181215e-06,
      "loss": 0.1685,
      "num_input_tokens_seen": 51223128,
      "step": 31850
    },
    {
      "epoch": 1.4420652346137306,
      "grad_norm": 0.8467779159545898,
      "learning_rate": 4.944401723840433e-06,
      "loss": 0.1197,
      "num_input_tokens_seen": 51230552,
      "step": 31855
    },
    {
      "epoch": 1.4422915865003736,
      "grad_norm": 0.5208131074905396,
      "learning_rate": 4.938542008324942e-06,
      "loss": 0.1417,
      "num_input_tokens_seen": 51238488,
      "step": 31860
    },
    {
      "epoch": 1.4425179383870166,
      "grad_norm": 0.9743098616600037,
      "learning_rate": 4.9326853865383855e-06,
      "loss": 0.1437,
      "num_input_tokens_seen": 51246584,
      "step": 31865
    },
    {
      "epoch": 1.4427442902736594,
      "grad_norm": 0.787969708442688,
      "learning_rate": 4.926831859383918e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 51254744,
      "step": 31870
    },
    {
      "epoch": 1.4429706421603024,
      "grad_norm": 1.211302638053894,
      "learning_rate": 4.92098142776424e-06,
      "loss": 0.1772,
      "num_input_tokens_seen": 51262552,
      "step": 31875
    },
    {
      "epoch": 1.4431969940469453,
      "grad_norm": 0.3401881754398346,
      "learning_rate": 4.91513409258155e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 51270552,
      "step": 31880
    },
    {
      "epoch": 1.4434233459335883,
      "grad_norm": 0.5685421824455261,
      "learning_rate": 4.909289854737581e-06,
      "loss": 0.1327,
      "num_input_tokens_seen": 51278136,
      "step": 31885
    },
    {
      "epoch": 1.4436496978202313,
      "grad_norm": 0.4984988570213318,
      "learning_rate": 4.903448715133602e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 51286072,
      "step": 31890
    },
    {
      "epoch": 1.4438760497068743,
      "grad_norm": 0.371121346950531,
      "learning_rate": 4.897610674670372e-06,
      "loss": 0.1856,
      "num_input_tokens_seen": 51294328,
      "step": 31895
    },
    {
      "epoch": 1.4441024015935173,
      "grad_norm": 0.5358260273933411,
      "learning_rate": 4.8917757342482e-06,
      "loss": 0.1482,
      "num_input_tokens_seen": 51302392,
      "step": 31900
    },
    {
      "epoch": 1.4443287534801603,
      "grad_norm": 0.3660917282104492,
      "learning_rate": 4.885943894766909e-06,
      "loss": 0.1013,
      "num_input_tokens_seen": 51310744,
      "step": 31905
    },
    {
      "epoch": 1.444555105366803,
      "grad_norm": 0.5065653920173645,
      "learning_rate": 4.880115157125842e-06,
      "loss": 0.1786,
      "num_input_tokens_seen": 51318840,
      "step": 31910
    },
    {
      "epoch": 1.444781457253446,
      "grad_norm": 0.618468701839447,
      "learning_rate": 4.874289522223857e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 51326712,
      "step": 31915
    },
    {
      "epoch": 1.445007809140089,
      "grad_norm": 1.2161469459533691,
      "learning_rate": 4.868466990959339e-06,
      "loss": 0.1791,
      "num_input_tokens_seen": 51334328,
      "step": 31920
    },
    {
      "epoch": 1.445234161026732,
      "grad_norm": 0.5937997102737427,
      "learning_rate": 4.8626475642301964e-06,
      "loss": 0.1875,
      "num_input_tokens_seen": 51342776,
      "step": 31925
    },
    {
      "epoch": 1.445460512913375,
      "grad_norm": 0.4182170331478119,
      "learning_rate": 4.856831242933871e-06,
      "loss": 0.139,
      "num_input_tokens_seen": 51350360,
      "step": 31930
    },
    {
      "epoch": 1.445686864800018,
      "grad_norm": 0.38453537225723267,
      "learning_rate": 4.851018027967294e-06,
      "loss": 0.167,
      "num_input_tokens_seen": 51358584,
      "step": 31935
    },
    {
      "epoch": 1.445913216686661,
      "grad_norm": 0.5095954537391663,
      "learning_rate": 4.845207920226946e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 51366680,
      "step": 31940
    },
    {
      "epoch": 1.446139568573304,
      "grad_norm": 0.8668342232704163,
      "learning_rate": 4.839400920608825e-06,
      "loss": 0.1133,
      "num_input_tokens_seen": 51375224,
      "step": 31945
    },
    {
      "epoch": 1.446365920459947,
      "grad_norm": 0.659085750579834,
      "learning_rate": 4.83359703000843e-06,
      "loss": 0.1475,
      "num_input_tokens_seen": 51383480,
      "step": 31950
    },
    {
      "epoch": 1.44659227234659,
      "grad_norm": 0.3099544942378998,
      "learning_rate": 4.827796249320804e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 51391128,
      "step": 31955
    },
    {
      "epoch": 1.446818624233233,
      "grad_norm": 0.32029807567596436,
      "learning_rate": 4.82199857944049e-06,
      "loss": 0.1172,
      "num_input_tokens_seen": 51398648,
      "step": 31960
    },
    {
      "epoch": 1.447044976119876,
      "grad_norm": 0.333258718252182,
      "learning_rate": 4.8162040212615695e-06,
      "loss": 0.1375,
      "num_input_tokens_seen": 51406776,
      "step": 31965
    },
    {
      "epoch": 1.447271328006519,
      "grad_norm": 0.9153702855110168,
      "learning_rate": 4.810412575677639e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 51414776,
      "step": 31970
    },
    {
      "epoch": 1.447497679893162,
      "grad_norm": 1.6719446182250977,
      "learning_rate": 4.804624243581801e-06,
      "loss": 0.1559,
      "num_input_tokens_seen": 51422424,
      "step": 31975
    },
    {
      "epoch": 1.447724031779805,
      "grad_norm": 0.6868846416473389,
      "learning_rate": 4.798839025866703e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 51430328,
      "step": 31980
    },
    {
      "epoch": 1.4479503836664478,
      "grad_norm": 0.42815542221069336,
      "learning_rate": 4.793056923424491e-06,
      "loss": 0.1417,
      "num_input_tokens_seen": 51437912,
      "step": 31985
    },
    {
      "epoch": 1.4481767355530908,
      "grad_norm": 0.8883547186851501,
      "learning_rate": 4.78727793714683e-06,
      "loss": 0.1412,
      "num_input_tokens_seen": 51446104,
      "step": 31990
    },
    {
      "epoch": 1.4484030874397338,
      "grad_norm": 0.4790915846824646,
      "learning_rate": 4.7815020679249285e-06,
      "loss": 0.132,
      "num_input_tokens_seen": 51454424,
      "step": 31995
    },
    {
      "epoch": 1.4486294393263768,
      "grad_norm": 0.8368338942527771,
      "learning_rate": 4.775729316649483e-06,
      "loss": 0.1366,
      "num_input_tokens_seen": 51462648,
      "step": 32000
    },
    {
      "epoch": 1.4486294393263768,
      "eval_loss": 0.14368723332881927,
      "eval_runtime": 404.3302,
      "eval_samples_per_second": 97.126,
      "eval_steps_per_second": 24.282,
      "num_input_tokens_seen": 51462648,
      "step": 32000
    },
    {
      "epoch": 1.4488557912130198,
      "grad_norm": 0.6018581390380859,
      "learning_rate": 4.769959684210728e-06,
      "loss": 0.1235,
      "num_input_tokens_seen": 51470712,
      "step": 32005
    },
    {
      "epoch": 1.4490821430996628,
      "grad_norm": 0.37030160427093506,
      "learning_rate": 4.764193171498426e-06,
      "loss": 0.092,
      "num_input_tokens_seen": 51479512,
      "step": 32010
    },
    {
      "epoch": 1.4493084949863058,
      "grad_norm": 0.953353762626648,
      "learning_rate": 4.75842977940183e-06,
      "loss": 0.1329,
      "num_input_tokens_seen": 51487768,
      "step": 32015
    },
    {
      "epoch": 1.4495348468729488,
      "grad_norm": 0.5307847857475281,
      "learning_rate": 4.752669508809729e-06,
      "loss": 0.153,
      "num_input_tokens_seen": 51496664,
      "step": 32020
    },
    {
      "epoch": 1.4497611987595915,
      "grad_norm": 1.0532102584838867,
      "learning_rate": 4.746912360610445e-06,
      "loss": 0.1726,
      "num_input_tokens_seen": 51504824,
      "step": 32025
    },
    {
      "epoch": 1.4499875506462345,
      "grad_norm": 0.8716606497764587,
      "learning_rate": 4.741158335691781e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 51513208,
      "step": 32030
    },
    {
      "epoch": 1.4502139025328775,
      "grad_norm": 0.616693913936615,
      "learning_rate": 4.7354074349410994e-06,
      "loss": 0.1461,
      "num_input_tokens_seen": 51521144,
      "step": 32035
    },
    {
      "epoch": 1.4504402544195205,
      "grad_norm": 0.6599060893058777,
      "learning_rate": 4.729659659245245e-06,
      "loss": 0.1425,
      "num_input_tokens_seen": 51529432,
      "step": 32040
    },
    {
      "epoch": 1.4506666063061635,
      "grad_norm": 0.35832479596138,
      "learning_rate": 4.723915009490601e-06,
      "loss": 0.1627,
      "num_input_tokens_seen": 51537112,
      "step": 32045
    },
    {
      "epoch": 1.4508929581928065,
      "grad_norm": 0.5292607545852661,
      "learning_rate": 4.718173486563077e-06,
      "loss": 0.1296,
      "num_input_tokens_seen": 51545464,
      "step": 32050
    },
    {
      "epoch": 1.4511193100794495,
      "grad_norm": 0.19433772563934326,
      "learning_rate": 4.71243509134808e-06,
      "loss": 0.0978,
      "num_input_tokens_seen": 51553592,
      "step": 32055
    },
    {
      "epoch": 1.4513456619660925,
      "grad_norm": 0.34923532605171204,
      "learning_rate": 4.706699824730532e-06,
      "loss": 0.1402,
      "num_input_tokens_seen": 51561560,
      "step": 32060
    },
    {
      "epoch": 1.4515720138527355,
      "grad_norm": 0.4902525544166565,
      "learning_rate": 4.700967687594901e-06,
      "loss": 0.1116,
      "num_input_tokens_seen": 51569528,
      "step": 32065
    },
    {
      "epoch": 1.4517983657393785,
      "grad_norm": 0.42666882276535034,
      "learning_rate": 4.69523868082514e-06,
      "loss": 0.1694,
      "num_input_tokens_seen": 51577752,
      "step": 32070
    },
    {
      "epoch": 1.4520247176260215,
      "grad_norm": 0.5343594551086426,
      "learning_rate": 4.689512805304747e-06,
      "loss": 0.1117,
      "num_input_tokens_seen": 51585848,
      "step": 32075
    },
    {
      "epoch": 1.4522510695126645,
      "grad_norm": 0.5448687672615051,
      "learning_rate": 4.683790061916707e-06,
      "loss": 0.1155,
      "num_input_tokens_seen": 51594488,
      "step": 32080
    },
    {
      "epoch": 1.4524774213993075,
      "grad_norm": 0.4081484377384186,
      "learning_rate": 4.678070451543551e-06,
      "loss": 0.1463,
      "num_input_tokens_seen": 51602808,
      "step": 32085
    },
    {
      "epoch": 1.4527037732859505,
      "grad_norm": 0.8173016905784607,
      "learning_rate": 4.6723539750673204e-06,
      "loss": 0.1372,
      "num_input_tokens_seen": 51611192,
      "step": 32090
    },
    {
      "epoch": 1.4529301251725932,
      "grad_norm": 0.6809319853782654,
      "learning_rate": 4.666640633369551e-06,
      "loss": 0.1402,
      "num_input_tokens_seen": 51619512,
      "step": 32095
    },
    {
      "epoch": 1.4531564770592362,
      "grad_norm": 0.7857496738433838,
      "learning_rate": 4.660930427331323e-06,
      "loss": 0.1241,
      "num_input_tokens_seen": 51627128,
      "step": 32100
    },
    {
      "epoch": 1.4533828289458792,
      "grad_norm": 1.125372290611267,
      "learning_rate": 4.6552233578332244e-06,
      "loss": 0.1592,
      "num_input_tokens_seen": 51635224,
      "step": 32105
    },
    {
      "epoch": 1.4536091808325222,
      "grad_norm": 0.5587084293365479,
      "learning_rate": 4.649519425755347e-06,
      "loss": 0.1282,
      "num_input_tokens_seen": 51643320,
      "step": 32110
    },
    {
      "epoch": 1.4538355327191652,
      "grad_norm": 1.1620993614196777,
      "learning_rate": 4.64381863197732e-06,
      "loss": 0.1814,
      "num_input_tokens_seen": 51651064,
      "step": 32115
    },
    {
      "epoch": 1.4540618846058082,
      "grad_norm": 1.1170904636383057,
      "learning_rate": 4.638120977378269e-06,
      "loss": 0.1633,
      "num_input_tokens_seen": 51658712,
      "step": 32120
    },
    {
      "epoch": 1.4542882364924512,
      "grad_norm": 0.5942880511283875,
      "learning_rate": 4.632426462836848e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 51667000,
      "step": 32125
    },
    {
      "epoch": 1.4545145883790942,
      "grad_norm": 0.41565749049186707,
      "learning_rate": 4.626735089231224e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 51674872,
      "step": 32130
    },
    {
      "epoch": 1.4547409402657372,
      "grad_norm": 0.6269365549087524,
      "learning_rate": 4.621046857439068e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 51682616,
      "step": 32135
    },
    {
      "epoch": 1.45496729215238,
      "grad_norm": 0.8211114406585693,
      "learning_rate": 4.615361768337587e-06,
      "loss": 0.1432,
      "num_input_tokens_seen": 51690712,
      "step": 32140
    },
    {
      "epoch": 1.455193644039023,
      "grad_norm": 0.7258415222167969,
      "learning_rate": 4.6096798228034946e-06,
      "loss": 0.1332,
      "num_input_tokens_seen": 51699384,
      "step": 32145
    },
    {
      "epoch": 1.455419995925666,
      "grad_norm": 0.7804009318351746,
      "learning_rate": 4.604001021713008e-06,
      "loss": 0.1336,
      "num_input_tokens_seen": 51706904,
      "step": 32150
    },
    {
      "epoch": 1.455646347812309,
      "grad_norm": 0.7265636324882507,
      "learning_rate": 4.598325365941883e-06,
      "loss": 0.1645,
      "num_input_tokens_seen": 51715000,
      "step": 32155
    },
    {
      "epoch": 1.455872699698952,
      "grad_norm": 0.6191685795783997,
      "learning_rate": 4.5926528563653645e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 51723064,
      "step": 32160
    },
    {
      "epoch": 1.456099051585595,
      "grad_norm": 0.6907321810722351,
      "learning_rate": 4.5869834938582295e-06,
      "loss": 0.1392,
      "num_input_tokens_seen": 51731064,
      "step": 32165
    },
    {
      "epoch": 1.456325403472238,
      "grad_norm": 0.3089386820793152,
      "learning_rate": 4.581317279294772e-06,
      "loss": 0.1169,
      "num_input_tokens_seen": 51738872,
      "step": 32170
    },
    {
      "epoch": 1.456551755358881,
      "grad_norm": 0.4587952196598053,
      "learning_rate": 4.57565421354878e-06,
      "loss": 0.1058,
      "num_input_tokens_seen": 51747800,
      "step": 32175
    },
    {
      "epoch": 1.456778107245524,
      "grad_norm": 0.2986406683921814,
      "learning_rate": 4.569994297493579e-06,
      "loss": 0.1853,
      "num_input_tokens_seen": 51755704,
      "step": 32180
    },
    {
      "epoch": 1.457004459132167,
      "grad_norm": 0.47026526927948,
      "learning_rate": 4.564337532002002e-06,
      "loss": 0.1416,
      "num_input_tokens_seen": 51764440,
      "step": 32185
    },
    {
      "epoch": 1.45723081101881,
      "grad_norm": 0.45473209023475647,
      "learning_rate": 4.55868391794638e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 51772152,
      "step": 32190
    },
    {
      "epoch": 1.457457162905453,
      "grad_norm": 0.32665860652923584,
      "learning_rate": 4.553033456198588e-06,
      "loss": 0.1293,
      "num_input_tokens_seen": 51780568,
      "step": 32195
    },
    {
      "epoch": 1.457683514792096,
      "grad_norm": 0.48734408617019653,
      "learning_rate": 4.54738614762999e-06,
      "loss": 0.1234,
      "num_input_tokens_seen": 51788568,
      "step": 32200
    },
    {
      "epoch": 1.457683514792096,
      "eval_loss": 0.14466197788715363,
      "eval_runtime": 404.6738,
      "eval_samples_per_second": 97.044,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 51788568,
      "step": 32200
    },
    {
      "epoch": 1.457909866678739,
      "grad_norm": 0.4939468204975128,
      "learning_rate": 4.541741993111465e-06,
      "loss": 0.1466,
      "num_input_tokens_seen": 51796856,
      "step": 32205
    },
    {
      "epoch": 1.4581362185653817,
      "grad_norm": 0.6274851560592651,
      "learning_rate": 4.536100993513423e-06,
      "loss": 0.1359,
      "num_input_tokens_seen": 51804792,
      "step": 32210
    },
    {
      "epoch": 1.4583625704520247,
      "grad_norm": 0.7048099637031555,
      "learning_rate": 4.530463149705768e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 51813176,
      "step": 32215
    },
    {
      "epoch": 1.4585889223386677,
      "grad_norm": 0.9957073926925659,
      "learning_rate": 4.524828462557934e-06,
      "loss": 0.1481,
      "num_input_tokens_seen": 51821432,
      "step": 32220
    },
    {
      "epoch": 1.4588152742253107,
      "grad_norm": 1.2319000959396362,
      "learning_rate": 4.5191969329388625e-06,
      "loss": 0.1669,
      "num_input_tokens_seen": 51830072,
      "step": 32225
    },
    {
      "epoch": 1.4590416261119536,
      "grad_norm": 0.7066144943237305,
      "learning_rate": 4.5135685617169965e-06,
      "loss": 0.1069,
      "num_input_tokens_seen": 51837784,
      "step": 32230
    },
    {
      "epoch": 1.4592679779985966,
      "grad_norm": 1.3147133588790894,
      "learning_rate": 4.507943349760313e-06,
      "loss": 0.1896,
      "num_input_tokens_seen": 51846296,
      "step": 32235
    },
    {
      "epoch": 1.4594943298852396,
      "grad_norm": 0.7779334783554077,
      "learning_rate": 4.502321297936277e-06,
      "loss": 0.1305,
      "num_input_tokens_seen": 51854616,
      "step": 32240
    },
    {
      "epoch": 1.4597206817718826,
      "grad_norm": 0.7741862535476685,
      "learning_rate": 4.496702407111888e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 51862136,
      "step": 32245
    },
    {
      "epoch": 1.4599470336585254,
      "grad_norm": 0.3643232583999634,
      "learning_rate": 4.491086678153653e-06,
      "loss": 0.1506,
      "num_input_tokens_seen": 51870136,
      "step": 32250
    },
    {
      "epoch": 1.4601733855451684,
      "grad_norm": 1.2678050994873047,
      "learning_rate": 4.485474111927579e-06,
      "loss": 0.1388,
      "num_input_tokens_seen": 51877592,
      "step": 32255
    },
    {
      "epoch": 1.4603997374318114,
      "grad_norm": 1.0309816598892212,
      "learning_rate": 4.479864709299197e-06,
      "loss": 0.1396,
      "num_input_tokens_seen": 51885624,
      "step": 32260
    },
    {
      "epoch": 1.4606260893184544,
      "grad_norm": 1.1568387746810913,
      "learning_rate": 4.474258471133555e-06,
      "loss": 0.1816,
      "num_input_tokens_seen": 51893336,
      "step": 32265
    },
    {
      "epoch": 1.4608524412050974,
      "grad_norm": 0.5141005516052246,
      "learning_rate": 4.4686553982952014e-06,
      "loss": 0.1572,
      "num_input_tokens_seen": 51901080,
      "step": 32270
    },
    {
      "epoch": 1.4610787930917404,
      "grad_norm": 0.7563551664352417,
      "learning_rate": 4.463055491648191e-06,
      "loss": 0.1277,
      "num_input_tokens_seen": 51908696,
      "step": 32275
    },
    {
      "epoch": 1.4613051449783834,
      "grad_norm": 0.4527375102043152,
      "learning_rate": 4.457458752056112e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 51916664,
      "step": 32280
    },
    {
      "epoch": 1.4615314968650264,
      "grad_norm": 0.615363359451294,
      "learning_rate": 4.451865180382042e-06,
      "loss": 0.1515,
      "num_input_tokens_seen": 51924216,
      "step": 32285
    },
    {
      "epoch": 1.4617578487516694,
      "grad_norm": 0.4225420355796814,
      "learning_rate": 4.4462747774885936e-06,
      "loss": 0.1382,
      "num_input_tokens_seen": 51932472,
      "step": 32290
    },
    {
      "epoch": 1.4619842006383124,
      "grad_norm": 0.7311407923698425,
      "learning_rate": 4.440687544237859e-06,
      "loss": 0.1806,
      "num_input_tokens_seen": 51940504,
      "step": 32295
    },
    {
      "epoch": 1.4622105525249554,
      "grad_norm": 0.7407963871955872,
      "learning_rate": 4.435103481491471e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 51948984,
      "step": 32300
    },
    {
      "epoch": 1.4624369044115983,
      "grad_norm": 1.0843840837478638,
      "learning_rate": 4.429522590110569e-06,
      "loss": 0.1364,
      "num_input_tokens_seen": 51957176,
      "step": 32305
    },
    {
      "epoch": 1.4626632562982413,
      "grad_norm": 0.7074148654937744,
      "learning_rate": 4.423944870955779e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 51965368,
      "step": 32310
    },
    {
      "epoch": 1.4628896081848843,
      "grad_norm": 2.5336191654205322,
      "learning_rate": 4.418370324887272e-06,
      "loss": 0.1958,
      "num_input_tokens_seen": 51973240,
      "step": 32315
    },
    {
      "epoch": 1.4631159600715273,
      "grad_norm": 0.5488117337226868,
      "learning_rate": 4.412798952764699e-06,
      "loss": 0.1452,
      "num_input_tokens_seen": 51981816,
      "step": 32320
    },
    {
      "epoch": 1.46334231195817,
      "grad_norm": 0.6255205869674683,
      "learning_rate": 4.407230755447245e-06,
      "loss": 0.1529,
      "num_input_tokens_seen": 51990328,
      "step": 32325
    },
    {
      "epoch": 1.463568663844813,
      "grad_norm": 0.3833425045013428,
      "learning_rate": 4.401665733793598e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 51997880,
      "step": 32330
    },
    {
      "epoch": 1.463795015731456,
      "grad_norm": 0.6033084988594055,
      "learning_rate": 4.3961038886619425e-06,
      "loss": 0.1264,
      "num_input_tokens_seen": 52005688,
      "step": 32335
    },
    {
      "epoch": 1.464021367618099,
      "grad_norm": 0.514218270778656,
      "learning_rate": 4.39054522091e-06,
      "loss": 0.131,
      "num_input_tokens_seen": 52013400,
      "step": 32340
    },
    {
      "epoch": 1.464247719504742,
      "grad_norm": 0.5878358483314514,
      "learning_rate": 4.384989731394979e-06,
      "loss": 0.1353,
      "num_input_tokens_seen": 52021400,
      "step": 32345
    },
    {
      "epoch": 1.464474071391385,
      "grad_norm": 0.9644140005111694,
      "learning_rate": 4.379437420973598e-06,
      "loss": 0.1823,
      "num_input_tokens_seen": 52028664,
      "step": 32350
    },
    {
      "epoch": 1.464700423278028,
      "grad_norm": 0.984904944896698,
      "learning_rate": 4.373888290502107e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 52036152,
      "step": 32355
    },
    {
      "epoch": 1.464926775164671,
      "grad_norm": 1.256941795349121,
      "learning_rate": 4.36834234083624e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 52043992,
      "step": 32360
    },
    {
      "epoch": 1.4651531270513138,
      "grad_norm": 0.4271032214164734,
      "learning_rate": 4.362799572831258e-06,
      "loss": 0.1654,
      "num_input_tokens_seen": 52051800,
      "step": 32365
    },
    {
      "epoch": 1.4653794789379568,
      "grad_norm": 1.1646648645401,
      "learning_rate": 4.35725998734193e-06,
      "loss": 0.1193,
      "num_input_tokens_seen": 52059800,
      "step": 32370
    },
    {
      "epoch": 1.4656058308245998,
      "grad_norm": 0.45120447874069214,
      "learning_rate": 4.3517235852225195e-06,
      "loss": 0.1291,
      "num_input_tokens_seen": 52067544,
      "step": 32375
    },
    {
      "epoch": 1.4658321827112428,
      "grad_norm": 0.5257042646408081,
      "learning_rate": 4.346190367326822e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 52075864,
      "step": 32380
    },
    {
      "epoch": 1.4660585345978858,
      "grad_norm": 0.777395486831665,
      "learning_rate": 4.340660334508115e-06,
      "loss": 0.1483,
      "num_input_tokens_seen": 52083832,
      "step": 32385
    },
    {
      "epoch": 1.4662848864845288,
      "grad_norm": 0.5921909213066101,
      "learning_rate": 4.335133487619206e-06,
      "loss": 0.1363,
      "num_input_tokens_seen": 52091224,
      "step": 32390
    },
    {
      "epoch": 1.4665112383711718,
      "grad_norm": 0.47126176953315735,
      "learning_rate": 4.329609827512409e-06,
      "loss": 0.1473,
      "num_input_tokens_seen": 52099160,
      "step": 32395
    },
    {
      "epoch": 1.4667375902578148,
      "grad_norm": 0.496825635433197,
      "learning_rate": 4.324089355039531e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 52107544,
      "step": 32400
    },
    {
      "epoch": 1.4667375902578148,
      "eval_loss": 0.1434798240661621,
      "eval_runtime": 404.7634,
      "eval_samples_per_second": 97.022,
      "eval_steps_per_second": 24.256,
      "num_input_tokens_seen": 52107544,
      "step": 32400
    },
    {
      "epoch": 1.4669639421444578,
      "grad_norm": 0.2853855788707733,
      "learning_rate": 4.3185720710519075e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 52115256,
      "step": 32405
    },
    {
      "epoch": 1.4671902940311008,
      "grad_norm": 1.0454167127609253,
      "learning_rate": 4.3130579764003724e-06,
      "loss": 0.1425,
      "num_input_tokens_seen": 52123032,
      "step": 32410
    },
    {
      "epoch": 1.4674166459177438,
      "grad_norm": 0.644935131072998,
      "learning_rate": 4.307547071935267e-06,
      "loss": 0.198,
      "num_input_tokens_seen": 52131608,
      "step": 32415
    },
    {
      "epoch": 1.4676429978043868,
      "grad_norm": 0.38801708817481995,
      "learning_rate": 4.302039358506435e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 52139384,
      "step": 32420
    },
    {
      "epoch": 1.4678693496910298,
      "grad_norm": 0.6013832092285156,
      "learning_rate": 4.296534836963245e-06,
      "loss": 0.1488,
      "num_input_tokens_seen": 52147992,
      "step": 32425
    },
    {
      "epoch": 1.4680957015776728,
      "grad_norm": 0.4078162610530853,
      "learning_rate": 4.291033508154555e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 52155640,
      "step": 32430
    },
    {
      "epoch": 1.4683220534643155,
      "grad_norm": 0.7396367192268372,
      "learning_rate": 4.285535372928748e-06,
      "loss": 0.1629,
      "num_input_tokens_seen": 52163512,
      "step": 32435
    },
    {
      "epoch": 1.4685484053509585,
      "grad_norm": 0.6605595946311951,
      "learning_rate": 4.280040432133695e-06,
      "loss": 0.0985,
      "num_input_tokens_seen": 52171736,
      "step": 32440
    },
    {
      "epoch": 1.4687747572376015,
      "grad_norm": 0.34118688106536865,
      "learning_rate": 4.274548686616789e-06,
      "loss": 0.1194,
      "num_input_tokens_seen": 52179672,
      "step": 32445
    },
    {
      "epoch": 1.4690011091242445,
      "grad_norm": 0.7447726726531982,
      "learning_rate": 4.2690601372249364e-06,
      "loss": 0.153,
      "num_input_tokens_seen": 52187800,
      "step": 32450
    },
    {
      "epoch": 1.4692274610108875,
      "grad_norm": 0.41862937808036804,
      "learning_rate": 4.263574784804525e-06,
      "loss": 0.1537,
      "num_input_tokens_seen": 52195672,
      "step": 32455
    },
    {
      "epoch": 1.4694538128975305,
      "grad_norm": 0.9356818199157715,
      "learning_rate": 4.258092630201479e-06,
      "loss": 0.1016,
      "num_input_tokens_seen": 52203672,
      "step": 32460
    },
    {
      "epoch": 1.4696801647841735,
      "grad_norm": 0.6246486902236938,
      "learning_rate": 4.252613674261202e-06,
      "loss": 0.1516,
      "num_input_tokens_seen": 52212376,
      "step": 32465
    },
    {
      "epoch": 1.4699065166708165,
      "grad_norm": 0.5641279220581055,
      "learning_rate": 4.2471379178286224e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 52220536,
      "step": 32470
    },
    {
      "epoch": 1.4701328685574595,
      "grad_norm": 0.7870048880577087,
      "learning_rate": 4.241665361748181e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 52228504,
      "step": 32475
    },
    {
      "epoch": 1.4703592204441023,
      "grad_norm": 0.7943083643913269,
      "learning_rate": 4.2361960068637994e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 52236344,
      "step": 32480
    },
    {
      "epoch": 1.4705855723307453,
      "grad_norm": 0.4767465591430664,
      "learning_rate": 4.230729854018933e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 52244984,
      "step": 32485
    },
    {
      "epoch": 1.4708119242173883,
      "grad_norm": 0.5919209122657776,
      "learning_rate": 4.225266904056521e-06,
      "loss": 0.1692,
      "num_input_tokens_seen": 52253368,
      "step": 32490
    },
    {
      "epoch": 1.4710382761040313,
      "grad_norm": 0.7348583340644836,
      "learning_rate": 4.21980715781903e-06,
      "loss": 0.1396,
      "num_input_tokens_seen": 52261016,
      "step": 32495
    },
    {
      "epoch": 1.4712646279906743,
      "grad_norm": 0.21577188372612,
      "learning_rate": 4.214350616148416e-06,
      "loss": 0.1486,
      "num_input_tokens_seen": 52269880,
      "step": 32500
    },
    {
      "epoch": 1.4714909798773173,
      "grad_norm": 0.8899747729301453,
      "learning_rate": 4.20889727988614e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 52277432,
      "step": 32505
    },
    {
      "epoch": 1.4717173317639602,
      "grad_norm": 0.815052330493927,
      "learning_rate": 4.20344714987318e-06,
      "loss": 0.136,
      "num_input_tokens_seen": 52285464,
      "step": 32510
    },
    {
      "epoch": 1.4719436836506032,
      "grad_norm": 0.44924432039260864,
      "learning_rate": 4.198000226950022e-06,
      "loss": 0.1622,
      "num_input_tokens_seen": 52293464,
      "step": 32515
    },
    {
      "epoch": 1.4721700355372462,
      "grad_norm": 0.3517358601093292,
      "learning_rate": 4.192556511956635e-06,
      "loss": 0.1214,
      "num_input_tokens_seen": 52301240,
      "step": 32520
    },
    {
      "epoch": 1.4723963874238892,
      "grad_norm": 0.5918239951133728,
      "learning_rate": 4.18711600573252e-06,
      "loss": 0.1943,
      "num_input_tokens_seen": 52309368,
      "step": 32525
    },
    {
      "epoch": 1.4726227393105322,
      "grad_norm": 0.5097747445106506,
      "learning_rate": 4.181678709116671e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 52317144,
      "step": 32530
    },
    {
      "epoch": 1.4728490911971752,
      "grad_norm": 0.7444620728492737,
      "learning_rate": 4.1762446229475785e-06,
      "loss": 0.0956,
      "num_input_tokens_seen": 52324888,
      "step": 32535
    },
    {
      "epoch": 1.4730754430838182,
      "grad_norm": 0.4477534294128418,
      "learning_rate": 4.17081374806326e-06,
      "loss": 0.1119,
      "num_input_tokens_seen": 52332920,
      "step": 32540
    },
    {
      "epoch": 1.4733017949704612,
      "grad_norm": 0.43962791562080383,
      "learning_rate": 4.165386085301212e-06,
      "loss": 0.1628,
      "num_input_tokens_seen": 52341176,
      "step": 32545
    },
    {
      "epoch": 1.473528146857104,
      "grad_norm": 0.4977375566959381,
      "learning_rate": 4.1599616354984525e-06,
      "loss": 0.1279,
      "num_input_tokens_seen": 52348504,
      "step": 32550
    },
    {
      "epoch": 1.473754498743747,
      "grad_norm": 1.3059139251708984,
      "learning_rate": 4.154540399491508e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 52356824,
      "step": 32555
    },
    {
      "epoch": 1.47398085063039,
      "grad_norm": 0.4761506915092468,
      "learning_rate": 4.149122378116394e-06,
      "loss": 0.1875,
      "num_input_tokens_seen": 52364600,
      "step": 32560
    },
    {
      "epoch": 1.474207202517033,
      "grad_norm": 0.4195447266101837,
      "learning_rate": 4.14370757220863e-06,
      "loss": 0.1312,
      "num_input_tokens_seen": 52372216,
      "step": 32565
    },
    {
      "epoch": 1.474433554403676,
      "grad_norm": 0.4086979329586029,
      "learning_rate": 4.138295982603263e-06,
      "loss": 0.1514,
      "num_input_tokens_seen": 52380664,
      "step": 32570
    },
    {
      "epoch": 1.474659906290319,
      "grad_norm": 0.6528878808021545,
      "learning_rate": 4.132887610134814e-06,
      "loss": 0.1228,
      "num_input_tokens_seen": 52388824,
      "step": 32575
    },
    {
      "epoch": 1.474886258176962,
      "grad_norm": 0.672267496585846,
      "learning_rate": 4.127482455637335e-06,
      "loss": 0.1973,
      "num_input_tokens_seen": 52396408,
      "step": 32580
    },
    {
      "epoch": 1.475112610063605,
      "grad_norm": 0.5347138047218323,
      "learning_rate": 4.1220805199443545e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 52405784,
      "step": 32585
    },
    {
      "epoch": 1.4753389619502477,
      "grad_norm": 0.6955483555793762,
      "learning_rate": 4.116681803888925e-06,
      "loss": 0.1408,
      "num_input_tokens_seen": 52414616,
      "step": 32590
    },
    {
      "epoch": 1.4755653138368907,
      "grad_norm": 0.6515504717826843,
      "learning_rate": 4.111286308303605e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 52422712,
      "step": 32595
    },
    {
      "epoch": 1.4757916657235337,
      "grad_norm": 0.6624001860618591,
      "learning_rate": 4.105894034020433e-06,
      "loss": 0.1197,
      "num_input_tokens_seen": 52430680,
      "step": 32600
    },
    {
      "epoch": 1.4757916657235337,
      "eval_loss": 0.14381857216358185,
      "eval_runtime": 405.1962,
      "eval_samples_per_second": 96.918,
      "eval_steps_per_second": 24.23,
      "num_input_tokens_seen": 52430680,
      "step": 32600
    },
    {
      "epoch": 1.4760180176101767,
      "grad_norm": 0.34315359592437744,
      "learning_rate": 4.100504981870975e-06,
      "loss": 0.0927,
      "num_input_tokens_seen": 52438616,
      "step": 32605
    },
    {
      "epoch": 1.4762443694968197,
      "grad_norm": 0.34438058733940125,
      "learning_rate": 4.0951191526862915e-06,
      "loss": 0.1344,
      "num_input_tokens_seen": 52446168,
      "step": 32610
    },
    {
      "epoch": 1.4764707213834627,
      "grad_norm": 0.8625017404556274,
      "learning_rate": 4.089736547296938e-06,
      "loss": 0.225,
      "num_input_tokens_seen": 52454296,
      "step": 32615
    },
    {
      "epoch": 1.4766970732701057,
      "grad_norm": 0.5837321281433105,
      "learning_rate": 4.08435716653299e-06,
      "loss": 0.1449,
      "num_input_tokens_seen": 52462104,
      "step": 32620
    },
    {
      "epoch": 1.4769234251567487,
      "grad_norm": 0.39294561743736267,
      "learning_rate": 4.0789810112240005e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 52470040,
      "step": 32625
    },
    {
      "epoch": 1.4771497770433917,
      "grad_norm": 0.8568999767303467,
      "learning_rate": 4.073608082199057e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 52477880,
      "step": 32630
    },
    {
      "epoch": 1.4773761289300347,
      "grad_norm": 1.184666395187378,
      "learning_rate": 4.068238380286718e-06,
      "loss": 0.1298,
      "num_input_tokens_seen": 52486360,
      "step": 32635
    },
    {
      "epoch": 1.4776024808166777,
      "grad_norm": 0.29279863834381104,
      "learning_rate": 4.062871906315072e-06,
      "loss": 0.1322,
      "num_input_tokens_seen": 52495256,
      "step": 32640
    },
    {
      "epoch": 1.4778288327033207,
      "grad_norm": 0.5164920091629028,
      "learning_rate": 4.057508661111686e-06,
      "loss": 0.1337,
      "num_input_tokens_seen": 52503480,
      "step": 32645
    },
    {
      "epoch": 1.4780551845899637,
      "grad_norm": 0.7657498121261597,
      "learning_rate": 4.052148645503648e-06,
      "loss": 0.1527,
      "num_input_tokens_seen": 52511928,
      "step": 32650
    },
    {
      "epoch": 1.4782815364766067,
      "grad_norm": 0.7385997772216797,
      "learning_rate": 4.046791860317531e-06,
      "loss": 0.1445,
      "num_input_tokens_seen": 52519416,
      "step": 32655
    },
    {
      "epoch": 1.4785078883632496,
      "grad_norm": 0.5394458770751953,
      "learning_rate": 4.041438306379431e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 52527160,
      "step": 32660
    },
    {
      "epoch": 1.4787342402498924,
      "grad_norm": 1.1701455116271973,
      "learning_rate": 4.036087984514916e-06,
      "loss": 0.1346,
      "num_input_tokens_seen": 52535672,
      "step": 32665
    },
    {
      "epoch": 1.4789605921365354,
      "grad_norm": 1.0922449827194214,
      "learning_rate": 4.030740895549084e-06,
      "loss": 0.1894,
      "num_input_tokens_seen": 52543672,
      "step": 32670
    },
    {
      "epoch": 1.4791869440231784,
      "grad_norm": 1.1079784631729126,
      "learning_rate": 4.025397040306531e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 52551416,
      "step": 32675
    },
    {
      "epoch": 1.4794132959098214,
      "grad_norm": 1.1743632555007935,
      "learning_rate": 4.0200564196113285e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 52559384,
      "step": 32680
    },
    {
      "epoch": 1.4796396477964644,
      "grad_norm": 0.6399680376052856,
      "learning_rate": 4.014719034287079e-06,
      "loss": 0.1087,
      "num_input_tokens_seen": 52568152,
      "step": 32685
    },
    {
      "epoch": 1.4798659996831074,
      "grad_norm": 0.5240419507026672,
      "learning_rate": 4.0093848851568775e-06,
      "loss": 0.1407,
      "num_input_tokens_seen": 52575960,
      "step": 32690
    },
    {
      "epoch": 1.4800923515697504,
      "grad_norm": 0.6523088216781616,
      "learning_rate": 4.004053973043304e-06,
      "loss": 0.1581,
      "num_input_tokens_seen": 52584024,
      "step": 32695
    },
    {
      "epoch": 1.4803187034563934,
      "grad_norm": 0.6417495608329773,
      "learning_rate": 3.998726298768465e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 52592504,
      "step": 32700
    },
    {
      "epoch": 1.4805450553430362,
      "grad_norm": 0.7338933348655701,
      "learning_rate": 3.99340186315395e-06,
      "loss": 0.1222,
      "num_input_tokens_seen": 52600184,
      "step": 32705
    },
    {
      "epoch": 1.4807714072296791,
      "grad_norm": 0.8119438290596008,
      "learning_rate": 3.988080667020849e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 52608344,
      "step": 32710
    },
    {
      "epoch": 1.4809977591163221,
      "grad_norm": 0.7473072409629822,
      "learning_rate": 3.982762711189766e-06,
      "loss": 0.1435,
      "num_input_tokens_seen": 52616024,
      "step": 32715
    },
    {
      "epoch": 1.4812241110029651,
      "grad_norm": 0.574582040309906,
      "learning_rate": 3.977447996480785e-06,
      "loss": 0.1285,
      "num_input_tokens_seen": 52624120,
      "step": 32720
    },
    {
      "epoch": 1.4814504628896081,
      "grad_norm": 0.4288485646247864,
      "learning_rate": 3.97213652371351e-06,
      "loss": 0.1774,
      "num_input_tokens_seen": 52631992,
      "step": 32725
    },
    {
      "epoch": 1.4816768147762511,
      "grad_norm": 0.4179283678531647,
      "learning_rate": 3.966828293707042e-06,
      "loss": 0.1318,
      "num_input_tokens_seen": 52639736,
      "step": 32730
    },
    {
      "epoch": 1.4819031666628941,
      "grad_norm": 0.26639842987060547,
      "learning_rate": 3.961523307279963e-06,
      "loss": 0.1396,
      "num_input_tokens_seen": 52648248,
      "step": 32735
    },
    {
      "epoch": 1.4821295185495371,
      "grad_norm": 1.3140344619750977,
      "learning_rate": 3.956221565250382e-06,
      "loss": 0.1512,
      "num_input_tokens_seen": 52656312,
      "step": 32740
    },
    {
      "epoch": 1.4823558704361801,
      "grad_norm": 0.4857769012451172,
      "learning_rate": 3.950923068435883e-06,
      "loss": 0.1183,
      "num_input_tokens_seen": 52663864,
      "step": 32745
    },
    {
      "epoch": 1.482582222322823,
      "grad_norm": 0.48509618639945984,
      "learning_rate": 3.945627817653566e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 52671832,
      "step": 32750
    },
    {
      "epoch": 1.482808574209466,
      "grad_norm": 0.40998587012290955,
      "learning_rate": 3.9403358137200335e-06,
      "loss": 0.1254,
      "num_input_tokens_seen": 52680408,
      "step": 32755
    },
    {
      "epoch": 1.483034926096109,
      "grad_norm": 0.6829438209533691,
      "learning_rate": 3.9350470574513605e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 52687960,
      "step": 32760
    },
    {
      "epoch": 1.483261277982752,
      "grad_norm": 0.6537967920303345,
      "learning_rate": 3.9297615496631525e-06,
      "loss": 0.1587,
      "num_input_tokens_seen": 52695640,
      "step": 32765
    },
    {
      "epoch": 1.483487629869395,
      "grad_norm": 0.6804336309432983,
      "learning_rate": 3.924479291170505e-06,
      "loss": 0.1868,
      "num_input_tokens_seen": 52703608,
      "step": 32770
    },
    {
      "epoch": 1.4837139817560379,
      "grad_norm": 0.6676579713821411,
      "learning_rate": 3.919200282788002e-06,
      "loss": 0.1765,
      "num_input_tokens_seen": 52711448,
      "step": 32775
    },
    {
      "epoch": 1.4839403336426809,
      "grad_norm": 0.31382253766059875,
      "learning_rate": 3.913924525329726e-06,
      "loss": 0.1446,
      "num_input_tokens_seen": 52719128,
      "step": 32780
    },
    {
      "epoch": 1.4841666855293238,
      "grad_norm": 0.6743339896202087,
      "learning_rate": 3.908652019609279e-06,
      "loss": 0.1484,
      "num_input_tokens_seen": 52726936,
      "step": 32785
    },
    {
      "epoch": 1.4843930374159668,
      "grad_norm": 0.551412045955658,
      "learning_rate": 3.9033827664397364e-06,
      "loss": 0.1578,
      "num_input_tokens_seen": 52734552,
      "step": 32790
    },
    {
      "epoch": 1.4846193893026098,
      "grad_norm": 0.38342371582984924,
      "learning_rate": 3.898116766633694e-06,
      "loss": 0.1457,
      "num_input_tokens_seen": 52742904,
      "step": 32795
    },
    {
      "epoch": 1.4848457411892528,
      "grad_norm": 0.37940463423728943,
      "learning_rate": 3.8928540210032225e-06,
      "loss": 0.1917,
      "num_input_tokens_seen": 52750648,
      "step": 32800
    },
    {
      "epoch": 1.4848457411892528,
      "eval_loss": 0.14405758678913116,
      "eval_runtime": 404.6234,
      "eval_samples_per_second": 97.056,
      "eval_steps_per_second": 24.265,
      "num_input_tokens_seen": 52750648,
      "step": 32800
    },
    {
      "epoch": 1.4850720930758958,
      "grad_norm": 0.891201376914978,
      "learning_rate": 3.887594530359909e-06,
      "loss": 0.1199,
      "num_input_tokens_seen": 52758040,
      "step": 32805
    },
    {
      "epoch": 1.4852984449625388,
      "grad_norm": 0.9726096391677856,
      "learning_rate": 3.88233829551484e-06,
      "loss": 0.2011,
      "num_input_tokens_seen": 52765784,
      "step": 32810
    },
    {
      "epoch": 1.4855247968491818,
      "grad_norm": 0.6280508637428284,
      "learning_rate": 3.877085317278581e-06,
      "loss": 0.1389,
      "num_input_tokens_seen": 52774264,
      "step": 32815
    },
    {
      "epoch": 1.4857511487358246,
      "grad_norm": 1.0255329608917236,
      "learning_rate": 3.87183559646122e-06,
      "loss": 0.1458,
      "num_input_tokens_seen": 52781944,
      "step": 32820
    },
    {
      "epoch": 1.4859775006224676,
      "grad_norm": 0.7297874093055725,
      "learning_rate": 3.866589133872317e-06,
      "loss": 0.1605,
      "num_input_tokens_seen": 52789240,
      "step": 32825
    },
    {
      "epoch": 1.4862038525091106,
      "grad_norm": 0.6858330368995667,
      "learning_rate": 3.861345930320948e-06,
      "loss": 0.1301,
      "num_input_tokens_seen": 52797144,
      "step": 32830
    },
    {
      "epoch": 1.4864302043957536,
      "grad_norm": 0.5270547866821289,
      "learning_rate": 3.856105986615688e-06,
      "loss": 0.1539,
      "num_input_tokens_seen": 52804408,
      "step": 32835
    },
    {
      "epoch": 1.4866565562823966,
      "grad_norm": 0.8773188591003418,
      "learning_rate": 3.850869303564589e-06,
      "loss": 0.1617,
      "num_input_tokens_seen": 52812824,
      "step": 32840
    },
    {
      "epoch": 1.4868829081690396,
      "grad_norm": 0.880287230014801,
      "learning_rate": 3.845635881975226e-06,
      "loss": 0.1312,
      "num_input_tokens_seen": 52820664,
      "step": 32845
    },
    {
      "epoch": 1.4871092600556826,
      "grad_norm": 0.3906010687351227,
      "learning_rate": 3.840405722654647e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 52828728,
      "step": 32850
    },
    {
      "epoch": 1.4873356119423256,
      "grad_norm": 0.6975807547569275,
      "learning_rate": 3.835178826409419e-06,
      "loss": 0.1635,
      "num_input_tokens_seen": 52836952,
      "step": 32855
    },
    {
      "epoch": 1.4875619638289685,
      "grad_norm": 0.3486131429672241,
      "learning_rate": 3.8299551940455895e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 52844920,
      "step": 32860
    },
    {
      "epoch": 1.4877883157156115,
      "grad_norm": 0.4284428656101227,
      "learning_rate": 3.824734826368703e-06,
      "loss": 0.1298,
      "num_input_tokens_seen": 52852632,
      "step": 32865
    },
    {
      "epoch": 1.4880146676022545,
      "grad_norm": 0.6320675611495972,
      "learning_rate": 3.819517724183813e-06,
      "loss": 0.1494,
      "num_input_tokens_seen": 52860984,
      "step": 32870
    },
    {
      "epoch": 1.4882410194888975,
      "grad_norm": 0.6731614470481873,
      "learning_rate": 3.8143038882954648e-06,
      "loss": 0.1222,
      "num_input_tokens_seen": 52869048,
      "step": 32875
    },
    {
      "epoch": 1.4884673713755405,
      "grad_norm": 0.5982881188392639,
      "learning_rate": 3.8090933195076867e-06,
      "loss": 0.1849,
      "num_input_tokens_seen": 52877592,
      "step": 32880
    },
    {
      "epoch": 1.4886937232621835,
      "grad_norm": 0.31944572925567627,
      "learning_rate": 3.8038860186240198e-06,
      "loss": 0.1086,
      "num_input_tokens_seen": 52885912,
      "step": 32885
    },
    {
      "epoch": 1.4889200751488263,
      "grad_norm": 1.2834575176239014,
      "learning_rate": 3.7986819864475026e-06,
      "loss": 0.1667,
      "num_input_tokens_seen": 52894232,
      "step": 32890
    },
    {
      "epoch": 1.4891464270354693,
      "grad_norm": 0.5988839268684387,
      "learning_rate": 3.793481223780651e-06,
      "loss": 0.1764,
      "num_input_tokens_seen": 52902584,
      "step": 32895
    },
    {
      "epoch": 1.4893727789221123,
      "grad_norm": 0.4868517220020294,
      "learning_rate": 3.788283731425496e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 52910936,
      "step": 32900
    },
    {
      "epoch": 1.4895991308087553,
      "grad_norm": 0.8322433233261108,
      "learning_rate": 3.7830895101835488e-06,
      "loss": 0.1818,
      "num_input_tokens_seen": 52919096,
      "step": 32905
    },
    {
      "epoch": 1.4898254826953983,
      "grad_norm": 1.2055165767669678,
      "learning_rate": 3.7778985608558274e-06,
      "loss": 0.2324,
      "num_input_tokens_seen": 52927192,
      "step": 32910
    },
    {
      "epoch": 1.4900518345820413,
      "grad_norm": 0.3748281002044678,
      "learning_rate": 3.7727108842428443e-06,
      "loss": 0.1464,
      "num_input_tokens_seen": 52934840,
      "step": 32915
    },
    {
      "epoch": 1.4902781864686843,
      "grad_norm": 0.6298511624336243,
      "learning_rate": 3.7675264811446065e-06,
      "loss": 0.1841,
      "num_input_tokens_seen": 52942744,
      "step": 32920
    },
    {
      "epoch": 1.4905045383553273,
      "grad_norm": 1.340701937675476,
      "learning_rate": 3.7623453523605994e-06,
      "loss": 0.1291,
      "num_input_tokens_seen": 52950328,
      "step": 32925
    },
    {
      "epoch": 1.49073089024197,
      "grad_norm": 0.9382152557373047,
      "learning_rate": 3.757167498689834e-06,
      "loss": 0.1539,
      "num_input_tokens_seen": 52958168,
      "step": 32930
    },
    {
      "epoch": 1.490957242128613,
      "grad_norm": 0.2746746838092804,
      "learning_rate": 3.7519929209307914e-06,
      "loss": 0.1115,
      "num_input_tokens_seen": 52966072,
      "step": 32935
    },
    {
      "epoch": 1.491183594015256,
      "grad_norm": 0.40247124433517456,
      "learning_rate": 3.746821619881463e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 52973560,
      "step": 32940
    },
    {
      "epoch": 1.491409945901899,
      "grad_norm": 1.1326866149902344,
      "learning_rate": 3.74165359633932e-06,
      "loss": 0.1401,
      "num_input_tokens_seen": 52981624,
      "step": 32945
    },
    {
      "epoch": 1.491636297788542,
      "grad_norm": 0.49419450759887695,
      "learning_rate": 3.736488851101341e-06,
      "loss": 0.1703,
      "num_input_tokens_seen": 52989816,
      "step": 32950
    },
    {
      "epoch": 1.491862649675185,
      "grad_norm": 0.5059596300125122,
      "learning_rate": 3.7313273849640035e-06,
      "loss": 0.1771,
      "num_input_tokens_seen": 52997304,
      "step": 32955
    },
    {
      "epoch": 1.492089001561828,
      "grad_norm": 0.577880859375,
      "learning_rate": 3.7261691987232533e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 53005848,
      "step": 32960
    },
    {
      "epoch": 1.492315353448471,
      "grad_norm": 0.6243116855621338,
      "learning_rate": 3.7210142931745575e-06,
      "loss": 0.1417,
      "num_input_tokens_seen": 53014040,
      "step": 32965
    },
    {
      "epoch": 1.492541705335114,
      "grad_norm": 0.9393545389175415,
      "learning_rate": 3.7158626691128712e-06,
      "loss": 0.124,
      "num_input_tokens_seen": 53022360,
      "step": 32970
    },
    {
      "epoch": 1.492768057221757,
      "grad_norm": 1.065305471420288,
      "learning_rate": 3.710714327332629e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 53030520,
      "step": 32975
    },
    {
      "epoch": 1.4929944091084,
      "grad_norm": 0.537743330001831,
      "learning_rate": 3.7055692686277815e-06,
      "loss": 0.142,
      "num_input_tokens_seen": 53038488,
      "step": 32980
    },
    {
      "epoch": 1.493220760995043,
      "grad_norm": 0.6451749205589294,
      "learning_rate": 3.70042749379175e-06,
      "loss": 0.1976,
      "num_input_tokens_seen": 53046040,
      "step": 32985
    },
    {
      "epoch": 1.493447112881686,
      "grad_norm": 0.5181066989898682,
      "learning_rate": 3.6952890036174693e-06,
      "loss": 0.1267,
      "num_input_tokens_seen": 53054008,
      "step": 32990
    },
    {
      "epoch": 1.493673464768329,
      "grad_norm": 0.6963859796524048,
      "learning_rate": 3.690153798897353e-06,
      "loss": 0.1701,
      "num_input_tokens_seen": 53062328,
      "step": 32995
    },
    {
      "epoch": 1.493899816654972,
      "grad_norm": 1.117875337600708,
      "learning_rate": 3.6850218804233225e-06,
      "loss": 0.179,
      "num_input_tokens_seen": 53070040,
      "step": 33000
    },
    {
      "epoch": 1.493899816654972,
      "eval_loss": 0.14386042952537537,
      "eval_runtime": 405.5335,
      "eval_samples_per_second": 96.838,
      "eval_steps_per_second": 24.21,
      "num_input_tokens_seen": 53070040,
      "step": 33000
    },
    {
      "epoch": 1.4941261685416147,
      "grad_norm": 0.6183487772941589,
      "learning_rate": 3.679893248986779e-06,
      "loss": 0.1592,
      "num_input_tokens_seen": 53078296,
      "step": 33005
    },
    {
      "epoch": 1.4943525204282577,
      "grad_norm": 0.5711138844490051,
      "learning_rate": 3.6747679053786147e-06,
      "loss": 0.1119,
      "num_input_tokens_seen": 53086808,
      "step": 33010
    },
    {
      "epoch": 1.4945788723149007,
      "grad_norm": 0.9408324956893921,
      "learning_rate": 3.669645850389228e-06,
      "loss": 0.1754,
      "num_input_tokens_seen": 53094616,
      "step": 33015
    },
    {
      "epoch": 1.4948052242015437,
      "grad_norm": 1.5204381942749023,
      "learning_rate": 3.664527084808514e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 53102136,
      "step": 33020
    },
    {
      "epoch": 1.4950315760881867,
      "grad_norm": 0.5688120126724243,
      "learning_rate": 3.6594116094258337e-06,
      "loss": 0.1466,
      "num_input_tokens_seen": 53109944,
      "step": 33025
    },
    {
      "epoch": 1.4952579279748297,
      "grad_norm": 0.9533811807632446,
      "learning_rate": 3.6542994250300665e-06,
      "loss": 0.154,
      "num_input_tokens_seen": 53117304,
      "step": 33030
    },
    {
      "epoch": 1.4954842798614727,
      "grad_norm": 1.1612944602966309,
      "learning_rate": 3.6491905324095825e-06,
      "loss": 0.1675,
      "num_input_tokens_seen": 53125304,
      "step": 33035
    },
    {
      "epoch": 1.4957106317481157,
      "grad_norm": 0.5060226917266846,
      "learning_rate": 3.644084932352221e-06,
      "loss": 0.1493,
      "num_input_tokens_seen": 53133016,
      "step": 33040
    },
    {
      "epoch": 1.4959369836347585,
      "grad_norm": 0.5726696848869324,
      "learning_rate": 3.6389826256453457e-06,
      "loss": 0.1229,
      "num_input_tokens_seen": 53141176,
      "step": 33045
    },
    {
      "epoch": 1.4961633355214015,
      "grad_norm": 0.5587347745895386,
      "learning_rate": 3.633883613075781e-06,
      "loss": 0.1592,
      "num_input_tokens_seen": 53149176,
      "step": 33050
    },
    {
      "epoch": 1.4963896874080445,
      "grad_norm": 0.7984314560890198,
      "learning_rate": 3.6287878954298693e-06,
      "loss": 0.1474,
      "num_input_tokens_seen": 53156856,
      "step": 33055
    },
    {
      "epoch": 1.4966160392946875,
      "grad_norm": 0.38087090849876404,
      "learning_rate": 3.6236954734934354e-06,
      "loss": 0.1417,
      "num_input_tokens_seen": 53165016,
      "step": 33060
    },
    {
      "epoch": 1.4968423911813304,
      "grad_norm": 0.6641663908958435,
      "learning_rate": 3.618606348051784e-06,
      "loss": 0.1433,
      "num_input_tokens_seen": 53173176,
      "step": 33065
    },
    {
      "epoch": 1.4970687430679734,
      "grad_norm": 0.43320733308792114,
      "learning_rate": 3.6135205198897376e-06,
      "loss": 0.1079,
      "num_input_tokens_seen": 53181432,
      "step": 33070
    },
    {
      "epoch": 1.4972950949546164,
      "grad_norm": 0.4730565845966339,
      "learning_rate": 3.6084379897915854e-06,
      "loss": 0.1534,
      "num_input_tokens_seen": 53189304,
      "step": 33075
    },
    {
      "epoch": 1.4975214468412594,
      "grad_norm": 0.4630798399448395,
      "learning_rate": 3.6033587585411115e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 53197784,
      "step": 33080
    },
    {
      "epoch": 1.4977477987279024,
      "grad_norm": 0.6572787165641785,
      "learning_rate": 3.5982828269216117e-06,
      "loss": 0.151,
      "num_input_tokens_seen": 53205784,
      "step": 33085
    },
    {
      "epoch": 1.4979741506145454,
      "grad_norm": 0.7344716787338257,
      "learning_rate": 3.593210195715843e-06,
      "loss": 0.1658,
      "num_input_tokens_seen": 53213496,
      "step": 33090
    },
    {
      "epoch": 1.4982005025011884,
      "grad_norm": 0.5564951300621033,
      "learning_rate": 3.5881408657060773e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 53220984,
      "step": 33095
    },
    {
      "epoch": 1.4984268543878314,
      "grad_norm": 0.5279649496078491,
      "learning_rate": 3.583074837674075e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 53229016,
      "step": 33100
    },
    {
      "epoch": 1.4986532062744744,
      "grad_norm": 0.6681100726127625,
      "learning_rate": 3.578012112401069e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 53237144,
      "step": 33105
    },
    {
      "epoch": 1.4988795581611174,
      "grad_norm": 0.7434732913970947,
      "learning_rate": 3.5729526906677996e-06,
      "loss": 0.1828,
      "num_input_tokens_seen": 53245816,
      "step": 33110
    },
    {
      "epoch": 1.4991059100477602,
      "grad_norm": 0.46191978454589844,
      "learning_rate": 3.5678965732545007e-06,
      "loss": 0.1328,
      "num_input_tokens_seen": 53254104,
      "step": 33115
    },
    {
      "epoch": 1.4993322619344032,
      "grad_norm": 0.9941295981407166,
      "learning_rate": 3.562843760940876e-06,
      "loss": 0.1257,
      "num_input_tokens_seen": 53262104,
      "step": 33120
    },
    {
      "epoch": 1.4995586138210462,
      "grad_norm": 0.36488965153694153,
      "learning_rate": 3.5577942545061473e-06,
      "loss": 0.1235,
      "num_input_tokens_seen": 53270072,
      "step": 33125
    },
    {
      "epoch": 1.4997849657076892,
      "grad_norm": 0.5204833149909973,
      "learning_rate": 3.5527480547289967e-06,
      "loss": 0.0881,
      "num_input_tokens_seen": 53277752,
      "step": 33130
    },
    {
      "epoch": 1.5000113175943321,
      "grad_norm": 0.9180947542190552,
      "learning_rate": 3.547705162387624e-06,
      "loss": 0.1548,
      "num_input_tokens_seen": 53285816,
      "step": 33135
    },
    {
      "epoch": 1.5002376694809751,
      "grad_norm": 0.7468688488006592,
      "learning_rate": 3.542665578259699e-06,
      "loss": 0.161,
      "num_input_tokens_seen": 53293624,
      "step": 33140
    },
    {
      "epoch": 1.5004640213676181,
      "grad_norm": 1.5046275854110718,
      "learning_rate": 3.5376293031223945e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 53301880,
      "step": 33145
    },
    {
      "epoch": 1.500690373254261,
      "grad_norm": 0.6514313817024231,
      "learning_rate": 3.5325963377523614e-06,
      "loss": 0.1264,
      "num_input_tokens_seen": 53310840,
      "step": 33150
    },
    {
      "epoch": 1.500916725140904,
      "grad_norm": 1.0801621675491333,
      "learning_rate": 3.5275666829257536e-06,
      "loss": 0.16,
      "num_input_tokens_seen": 53319000,
      "step": 33155
    },
    {
      "epoch": 1.501143077027547,
      "grad_norm": 0.8889193534851074,
      "learning_rate": 3.5225403394181955e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 53326680,
      "step": 33160
    },
    {
      "epoch": 1.50136942891419,
      "grad_norm": 0.5114122033119202,
      "learning_rate": 3.517517308004828e-06,
      "loss": 0.1201,
      "num_input_tokens_seen": 53334936,
      "step": 33165
    },
    {
      "epoch": 1.501595780800833,
      "grad_norm": 0.4983021914958954,
      "learning_rate": 3.512497589460251e-06,
      "loss": 0.1303,
      "num_input_tokens_seen": 53342808,
      "step": 33170
    },
    {
      "epoch": 1.5018221326874759,
      "grad_norm": 0.6590192914009094,
      "learning_rate": 3.5074811845585727e-06,
      "loss": 0.1602,
      "num_input_tokens_seen": 53351032,
      "step": 33175
    },
    {
      "epoch": 1.5020484845741189,
      "grad_norm": 0.721049427986145,
      "learning_rate": 3.5024680940733937e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 53359768,
      "step": 33180
    },
    {
      "epoch": 1.5022748364607619,
      "grad_norm": 1.4617195129394531,
      "learning_rate": 3.4974583187777852e-06,
      "loss": 0.1448,
      "num_input_tokens_seen": 53368216,
      "step": 33185
    },
    {
      "epoch": 1.5025011883474049,
      "grad_norm": 0.7417003512382507,
      "learning_rate": 3.4924518594443204e-06,
      "loss": 0.1462,
      "num_input_tokens_seen": 53376408,
      "step": 33190
    },
    {
      "epoch": 1.5027275402340479,
      "grad_norm": 0.7467273473739624,
      "learning_rate": 3.4874487168450682e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 53384440,
      "step": 33195
    },
    {
      "epoch": 1.5029538921206909,
      "grad_norm": 0.2836850881576538,
      "learning_rate": 3.482448891751558e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 53392248,
      "step": 33200
    },
    {
      "epoch": 1.5029538921206909,
      "eval_loss": 0.1439300924539566,
      "eval_runtime": 404.7541,
      "eval_samples_per_second": 97.024,
      "eval_steps_per_second": 24.257,
      "num_input_tokens_seen": 53392248,
      "step": 33200
    },
    {
      "epoch": 1.5031802440073339,
      "grad_norm": 0.9567021131515503,
      "learning_rate": 3.477452384934843e-06,
      "loss": 0.1408,
      "num_input_tokens_seen": 53400024,
      "step": 33205
    },
    {
      "epoch": 1.5034065958939768,
      "grad_norm": 1.699263095855713,
      "learning_rate": 3.472459197165434e-06,
      "loss": 0.1724,
      "num_input_tokens_seen": 53407992,
      "step": 33210
    },
    {
      "epoch": 1.5036329477806198,
      "grad_norm": 0.4214668869972229,
      "learning_rate": 3.4674693292133518e-06,
      "loss": 0.1118,
      "num_input_tokens_seen": 53415480,
      "step": 33215
    },
    {
      "epoch": 1.5038592996672628,
      "grad_norm": 0.5257816910743713,
      "learning_rate": 3.4624827818480977e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 53423448,
      "step": 33220
    },
    {
      "epoch": 1.5040856515539058,
      "grad_norm": 0.7173994183540344,
      "learning_rate": 3.4574995558386474e-06,
      "loss": 0.1643,
      "num_input_tokens_seen": 53431832,
      "step": 33225
    },
    {
      "epoch": 1.5043120034405488,
      "grad_norm": 0.6927089095115662,
      "learning_rate": 3.452519651953487e-06,
      "loss": 0.1416,
      "num_input_tokens_seen": 53439960,
      "step": 33230
    },
    {
      "epoch": 1.5045383553271916,
      "grad_norm": 0.3666647970676422,
      "learning_rate": 3.447543070960585e-06,
      "loss": 0.1511,
      "num_input_tokens_seen": 53448056,
      "step": 33235
    },
    {
      "epoch": 1.5047647072138346,
      "grad_norm": 0.5801924467086792,
      "learning_rate": 3.4425698136273778e-06,
      "loss": 0.1291,
      "num_input_tokens_seen": 53456760,
      "step": 33240
    },
    {
      "epoch": 1.5049910591004776,
      "grad_norm": 0.8267382979393005,
      "learning_rate": 3.437599880720821e-06,
      "loss": 0.1593,
      "num_input_tokens_seen": 53464760,
      "step": 33245
    },
    {
      "epoch": 1.5052174109871206,
      "grad_norm": 0.7421447038650513,
      "learning_rate": 3.4326332730073267e-06,
      "loss": 0.1152,
      "num_input_tokens_seen": 53473144,
      "step": 33250
    },
    {
      "epoch": 1.5054437628737636,
      "grad_norm": 0.6675805449485779,
      "learning_rate": 3.427669991252813e-06,
      "loss": 0.1256,
      "num_input_tokens_seen": 53480760,
      "step": 33255
    },
    {
      "epoch": 1.5056701147604066,
      "grad_norm": 0.22880695760250092,
      "learning_rate": 3.42271003622269e-06,
      "loss": 0.1465,
      "num_input_tokens_seen": 53488600,
      "step": 33260
    },
    {
      "epoch": 1.5058964666470493,
      "grad_norm": 1.0160523653030396,
      "learning_rate": 3.4177534086818286e-06,
      "loss": 0.1122,
      "num_input_tokens_seen": 53497112,
      "step": 33265
    },
    {
      "epoch": 1.5061228185336923,
      "grad_norm": 1.0101550817489624,
      "learning_rate": 3.412800109394612e-06,
      "loss": 0.1702,
      "num_input_tokens_seen": 53505912,
      "step": 33270
    },
    {
      "epoch": 1.5063491704203353,
      "grad_norm": 0.556938886642456,
      "learning_rate": 3.4078501391249044e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 53513432,
      "step": 33275
    },
    {
      "epoch": 1.5065755223069783,
      "grad_norm": 1.6838908195495605,
      "learning_rate": 3.4029034986360453e-06,
      "loss": 0.1732,
      "num_input_tokens_seen": 53522744,
      "step": 33280
    },
    {
      "epoch": 1.5068018741936213,
      "grad_norm": 0.42191311717033386,
      "learning_rate": 3.397960188690877e-06,
      "loss": 0.1334,
      "num_input_tokens_seen": 53530616,
      "step": 33285
    },
    {
      "epoch": 1.5070282260802643,
      "grad_norm": 0.6258224844932556,
      "learning_rate": 3.393020210051717e-06,
      "loss": 0.1272,
      "num_input_tokens_seen": 53538264,
      "step": 33290
    },
    {
      "epoch": 1.5072545779669073,
      "grad_norm": 0.45556414127349854,
      "learning_rate": 3.3880835634803655e-06,
      "loss": 0.153,
      "num_input_tokens_seen": 53546232,
      "step": 33295
    },
    {
      "epoch": 1.5074809298535503,
      "grad_norm": 0.5010591745376587,
      "learning_rate": 3.383150249738126e-06,
      "loss": 0.148,
      "num_input_tokens_seen": 53554712,
      "step": 33300
    },
    {
      "epoch": 1.5077072817401933,
      "grad_norm": 0.36429786682128906,
      "learning_rate": 3.3782202695857663e-06,
      "loss": 0.1776,
      "num_input_tokens_seen": 53562392,
      "step": 33305
    },
    {
      "epoch": 1.5079336336268363,
      "grad_norm": 0.48076778650283813,
      "learning_rate": 3.373293623783558e-06,
      "loss": 0.1345,
      "num_input_tokens_seen": 53570232,
      "step": 33310
    },
    {
      "epoch": 1.5081599855134793,
      "grad_norm": 0.439363032579422,
      "learning_rate": 3.368370313091257e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 53578136,
      "step": 33315
    },
    {
      "epoch": 1.5083863374001223,
      "grad_norm": 0.41220682859420776,
      "learning_rate": 3.363450338268087e-06,
      "loss": 0.0977,
      "num_input_tokens_seen": 53586872,
      "step": 33320
    },
    {
      "epoch": 1.5086126892867653,
      "grad_norm": 1.3764936923980713,
      "learning_rate": 3.358533700072783e-06,
      "loss": 0.1538,
      "num_input_tokens_seen": 53594968,
      "step": 33325
    },
    {
      "epoch": 1.5088390411734083,
      "grad_norm": 0.45528465509414673,
      "learning_rate": 3.3536203992635377e-06,
      "loss": 0.1098,
      "num_input_tokens_seen": 53602648,
      "step": 33330
    },
    {
      "epoch": 1.5090653930600513,
      "grad_norm": 0.5608273148536682,
      "learning_rate": 3.348710436598057e-06,
      "loss": 0.1576,
      "num_input_tokens_seen": 53610712,
      "step": 33335
    },
    {
      "epoch": 1.5092917449466943,
      "grad_norm": 0.35186994075775146,
      "learning_rate": 3.3438038128335155e-06,
      "loss": 0.1408,
      "num_input_tokens_seen": 53618776,
      "step": 33340
    },
    {
      "epoch": 1.5095180968333373,
      "grad_norm": 0.6466678977012634,
      "learning_rate": 3.338900528726571e-06,
      "loss": 0.1484,
      "num_input_tokens_seen": 53627160,
      "step": 33345
    },
    {
      "epoch": 1.50974444871998,
      "grad_norm": 0.6847141981124878,
      "learning_rate": 3.3340005850333812e-06,
      "loss": 0.1694,
      "num_input_tokens_seen": 53635704,
      "step": 33350
    },
    {
      "epoch": 1.509970800606623,
      "grad_norm": 0.43914327025413513,
      "learning_rate": 3.329103982509568e-06,
      "loss": 0.1174,
      "num_input_tokens_seen": 53643320,
      "step": 33355
    },
    {
      "epoch": 1.510197152493266,
      "grad_norm": 1.249948501586914,
      "learning_rate": 3.324210721910259e-06,
      "loss": 0.1555,
      "num_input_tokens_seen": 53651288,
      "step": 33360
    },
    {
      "epoch": 1.510423504379909,
      "grad_norm": 0.2709967792034149,
      "learning_rate": 3.319320803990053e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 53659160,
      "step": 33365
    },
    {
      "epoch": 1.510649856266552,
      "grad_norm": 0.8095479011535645,
      "learning_rate": 3.3144342295030274e-06,
      "loss": 0.1579,
      "num_input_tokens_seen": 53666712,
      "step": 33370
    },
    {
      "epoch": 1.510876208153195,
      "grad_norm": 0.5842145085334778,
      "learning_rate": 3.309550999202765e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 53674648,
      "step": 33375
    },
    {
      "epoch": 1.5111025600398378,
      "grad_norm": 0.37810853123664856,
      "learning_rate": 3.3046711138423197e-06,
      "loss": 0.1169,
      "num_input_tokens_seen": 53682520,
      "step": 33380
    },
    {
      "epoch": 1.5113289119264808,
      "grad_norm": 0.521538496017456,
      "learning_rate": 3.2997945741742255e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 53690776,
      "step": 33385
    },
    {
      "epoch": 1.5115552638131238,
      "grad_norm": 1.0715899467468262,
      "learning_rate": 3.2949213809505082e-06,
      "loss": 0.1386,
      "num_input_tokens_seen": 53699576,
      "step": 33390
    },
    {
      "epoch": 1.5117816156997668,
      "grad_norm": 0.6801825761795044,
      "learning_rate": 3.2900515349226834e-06,
      "loss": 0.1719,
      "num_input_tokens_seen": 53707384,
      "step": 33395
    },
    {
      "epoch": 1.5120079675864098,
      "grad_norm": 0.745128333568573,
      "learning_rate": 3.285185036841731e-06,
      "loss": 0.1338,
      "num_input_tokens_seen": 53715256,
      "step": 33400
    },
    {
      "epoch": 1.5120079675864098,
      "eval_loss": 0.1437237560749054,
      "eval_runtime": 404.7774,
      "eval_samples_per_second": 97.019,
      "eval_steps_per_second": 24.255,
      "num_input_tokens_seen": 53715256,
      "step": 33400
    },
    {
      "epoch": 1.5122343194730528,
      "grad_norm": 1.149015188217163,
      "learning_rate": 3.2803218874581377e-06,
      "loss": 0.1323,
      "num_input_tokens_seen": 53723032,
      "step": 33405
    },
    {
      "epoch": 1.5124606713596958,
      "grad_norm": 0.7135816216468811,
      "learning_rate": 3.2754620875218494e-06,
      "loss": 0.1263,
      "num_input_tokens_seen": 53731224,
      "step": 33410
    },
    {
      "epoch": 1.5126870232463387,
      "grad_norm": 0.453736275434494,
      "learning_rate": 3.2706056377823146e-06,
      "loss": 0.1525,
      "num_input_tokens_seen": 53738776,
      "step": 33415
    },
    {
      "epoch": 1.5129133751329817,
      "grad_norm": 0.6598725914955139,
      "learning_rate": 3.2657525389884647e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 53746648,
      "step": 33420
    },
    {
      "epoch": 1.5131397270196247,
      "grad_norm": 0.6064930558204651,
      "learning_rate": 3.260902791888698e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 53754520,
      "step": 33425
    },
    {
      "epoch": 1.5133660789062677,
      "grad_norm": 0.3887805640697479,
      "learning_rate": 3.2560563972309166e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 53762680,
      "step": 33430
    },
    {
      "epoch": 1.5135924307929107,
      "grad_norm": 0.7448542714118958,
      "learning_rate": 3.251213355762489e-06,
      "loss": 0.1205,
      "num_input_tokens_seen": 53770392,
      "step": 33435
    },
    {
      "epoch": 1.5138187826795537,
      "grad_norm": 0.7685848474502563,
      "learning_rate": 3.2463736682302707e-06,
      "loss": 0.1153,
      "num_input_tokens_seen": 53777944,
      "step": 33440
    },
    {
      "epoch": 1.5140451345661967,
      "grad_norm": 1.3324731588363647,
      "learning_rate": 3.2415373353806124e-06,
      "loss": 0.1747,
      "num_input_tokens_seen": 53786712,
      "step": 33445
    },
    {
      "epoch": 1.5142714864528397,
      "grad_norm": 1.271860122680664,
      "learning_rate": 3.236704357959322e-06,
      "loss": 0.127,
      "num_input_tokens_seen": 53795640,
      "step": 33450
    },
    {
      "epoch": 1.5144978383394827,
      "grad_norm": 0.6737464666366577,
      "learning_rate": 3.2318747367117154e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 53804024,
      "step": 33455
    },
    {
      "epoch": 1.5147241902261257,
      "grad_norm": 0.5002490878105164,
      "learning_rate": 3.227048472382585e-06,
      "loss": 0.1183,
      "num_input_tokens_seen": 53812440,
      "step": 33460
    },
    {
      "epoch": 1.5149505421127685,
      "grad_norm": 0.8094502687454224,
      "learning_rate": 3.2222255657161915e-06,
      "loss": 0.1891,
      "num_input_tokens_seen": 53820728,
      "step": 33465
    },
    {
      "epoch": 1.5151768939994115,
      "grad_norm": 0.4540242552757263,
      "learning_rate": 3.2174060174562924e-06,
      "loss": 0.1016,
      "num_input_tokens_seen": 53828536,
      "step": 33470
    },
    {
      "epoch": 1.5154032458860545,
      "grad_norm": 0.9747182726860046,
      "learning_rate": 3.2125898283461298e-06,
      "loss": 0.1497,
      "num_input_tokens_seen": 53836408,
      "step": 33475
    },
    {
      "epoch": 1.5156295977726975,
      "grad_norm": 0.3453916907310486,
      "learning_rate": 3.207776999128406e-06,
      "loss": 0.1112,
      "num_input_tokens_seen": 53843960,
      "step": 33480
    },
    {
      "epoch": 1.5158559496593405,
      "grad_norm": 0.5284923911094666,
      "learning_rate": 3.202967530545331e-06,
      "loss": 0.1523,
      "num_input_tokens_seen": 53851864,
      "step": 33485
    },
    {
      "epoch": 1.5160823015459832,
      "grad_norm": 0.44625306129455566,
      "learning_rate": 3.1981614233385778e-06,
      "loss": 0.1453,
      "num_input_tokens_seen": 53860440,
      "step": 33490
    },
    {
      "epoch": 1.5163086534326262,
      "grad_norm": 0.8299866914749146,
      "learning_rate": 3.1933586782493115e-06,
      "loss": 0.1776,
      "num_input_tokens_seen": 53868696,
      "step": 33495
    },
    {
      "epoch": 1.5165350053192692,
      "grad_norm": 0.5832003951072693,
      "learning_rate": 3.188559296018184e-06,
      "loss": 0.2031,
      "num_input_tokens_seen": 53876696,
      "step": 33500
    },
    {
      "epoch": 1.5167613572059122,
      "grad_norm": 0.6914666891098022,
      "learning_rate": 3.1837632773853098e-06,
      "loss": 0.1456,
      "num_input_tokens_seen": 53885048,
      "step": 33505
    },
    {
      "epoch": 1.5169877090925552,
      "grad_norm": 0.6644241809844971,
      "learning_rate": 3.178970623090294e-06,
      "loss": 0.1422,
      "num_input_tokens_seen": 53892504,
      "step": 33510
    },
    {
      "epoch": 1.5172140609791982,
      "grad_norm": 0.7016656994819641,
      "learning_rate": 3.174181333872234e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 53900696,
      "step": 33515
    },
    {
      "epoch": 1.5174404128658412,
      "grad_norm": 0.6322630643844604,
      "learning_rate": 3.169395410469686e-06,
      "loss": 0.1326,
      "num_input_tokens_seen": 53908408,
      "step": 33520
    },
    {
      "epoch": 1.5176667647524842,
      "grad_norm": 0.6007755994796753,
      "learning_rate": 3.164612853620713e-06,
      "loss": 0.1445,
      "num_input_tokens_seen": 53916440,
      "step": 33525
    },
    {
      "epoch": 1.5178931166391272,
      "grad_norm": 1.4688326120376587,
      "learning_rate": 3.1598336640628333e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 53924792,
      "step": 33530
    },
    {
      "epoch": 1.5181194685257702,
      "grad_norm": 1.841773271560669,
      "learning_rate": 3.155057842533063e-06,
      "loss": 0.1701,
      "num_input_tokens_seen": 53933080,
      "step": 33535
    },
    {
      "epoch": 1.5183458204124132,
      "grad_norm": 0.7429018020629883,
      "learning_rate": 3.1502853897678984e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 53940568,
      "step": 33540
    },
    {
      "epoch": 1.5185721722990562,
      "grad_norm": 0.7026719450950623,
      "learning_rate": 3.1455163065033017e-06,
      "loss": 0.1254,
      "num_input_tokens_seen": 53948248,
      "step": 33545
    },
    {
      "epoch": 1.5187985241856992,
      "grad_norm": 0.5371428728103638,
      "learning_rate": 3.140750593474734e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 53956536,
      "step": 33550
    },
    {
      "epoch": 1.5190248760723422,
      "grad_norm": 1.1537903547286987,
      "learning_rate": 3.1359882514171294e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 53965080,
      "step": 33555
    },
    {
      "epoch": 1.5192512279589852,
      "grad_norm": 0.7004197835922241,
      "learning_rate": 3.1312292810648903e-06,
      "loss": 0.1684,
      "num_input_tokens_seen": 53972344,
      "step": 33560
    },
    {
      "epoch": 1.5194775798456281,
      "grad_norm": 0.612489640712738,
      "learning_rate": 3.1264736831519204e-06,
      "loss": 0.1458,
      "num_input_tokens_seen": 53979768,
      "step": 33565
    },
    {
      "epoch": 1.5197039317322711,
      "grad_norm": 0.6170119047164917,
      "learning_rate": 3.1217214584115863e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 53987544,
      "step": 33570
    },
    {
      "epoch": 1.519930283618914,
      "grad_norm": 0.8094677925109863,
      "learning_rate": 3.116972607576746e-06,
      "loss": 0.1473,
      "num_input_tokens_seen": 53995320,
      "step": 33575
    },
    {
      "epoch": 1.520156635505557,
      "grad_norm": 0.7175721526145935,
      "learning_rate": 3.1122271313797303e-06,
      "loss": 0.1383,
      "num_input_tokens_seen": 54003192,
      "step": 33580
    },
    {
      "epoch": 1.5203829873922,
      "grad_norm": 0.3378582000732422,
      "learning_rate": 3.107485030552343e-06,
      "loss": 0.1633,
      "num_input_tokens_seen": 54010968,
      "step": 33585
    },
    {
      "epoch": 1.520609339278843,
      "grad_norm": 1.0840567350387573,
      "learning_rate": 3.1027463058258848e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 54019192,
      "step": 33590
    },
    {
      "epoch": 1.520835691165486,
      "grad_norm": 0.8721054196357727,
      "learning_rate": 3.0980109579311273e-06,
      "loss": 0.1042,
      "num_input_tokens_seen": 54027224,
      "step": 33595
    },
    {
      "epoch": 1.5210620430521289,
      "grad_norm": 0.8096022009849548,
      "learning_rate": 3.093278987598314e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 54035320,
      "step": 33600
    },
    {
      "epoch": 1.5210620430521289,
      "eval_loss": 0.14343371987342834,
      "eval_runtime": 405.2868,
      "eval_samples_per_second": 96.897,
      "eval_steps_per_second": 24.225,
      "num_input_tokens_seen": 54035320,
      "step": 33600
    },
    {
      "epoch": 1.5212883949387717,
      "grad_norm": 0.48062169551849365,
      "learning_rate": 3.0885503955571826e-06,
      "loss": 0.1724,
      "num_input_tokens_seen": 54043480,
      "step": 33605
    },
    {
      "epoch": 1.5215147468254147,
      "grad_norm": 0.42582806944847107,
      "learning_rate": 3.0838251825369313e-06,
      "loss": 0.1232,
      "num_input_tokens_seen": 54051640,
      "step": 33610
    },
    {
      "epoch": 1.5217410987120576,
      "grad_norm": 0.5807968378067017,
      "learning_rate": 3.0791033492662517e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 54059256,
      "step": 33615
    },
    {
      "epoch": 1.5219674505987006,
      "grad_norm": 0.3090733289718628,
      "learning_rate": 3.0743848964733203e-06,
      "loss": 0.1435,
      "num_input_tokens_seen": 54067544,
      "step": 33620
    },
    {
      "epoch": 1.5221938024853436,
      "grad_norm": 0.8120520710945129,
      "learning_rate": 3.0696698248857625e-06,
      "loss": 0.1583,
      "num_input_tokens_seen": 54075608,
      "step": 33625
    },
    {
      "epoch": 1.5224201543719866,
      "grad_norm": 0.9002156853675842,
      "learning_rate": 3.0649581352307192e-06,
      "loss": 0.1744,
      "num_input_tokens_seen": 54083608,
      "step": 33630
    },
    {
      "epoch": 1.5226465062586296,
      "grad_norm": 1.0045808553695679,
      "learning_rate": 3.060249828234776e-06,
      "loss": 0.1698,
      "num_input_tokens_seen": 54091448,
      "step": 33635
    },
    {
      "epoch": 1.5228728581452726,
      "grad_norm": 0.9508116245269775,
      "learning_rate": 3.055544904624025e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 54099896,
      "step": 33640
    },
    {
      "epoch": 1.5230992100319156,
      "grad_norm": 0.3029927909374237,
      "learning_rate": 3.050843365124026e-06,
      "loss": 0.1137,
      "num_input_tokens_seen": 54107672,
      "step": 33645
    },
    {
      "epoch": 1.5233255619185586,
      "grad_norm": 0.5393630266189575,
      "learning_rate": 3.0461452104598083e-06,
      "loss": 0.1088,
      "num_input_tokens_seen": 54115288,
      "step": 33650
    },
    {
      "epoch": 1.5235519138052016,
      "grad_norm": 0.46008065342903137,
      "learning_rate": 3.0414504413558836e-06,
      "loss": 0.1472,
      "num_input_tokens_seen": 54122968,
      "step": 33655
    },
    {
      "epoch": 1.5237782656918446,
      "grad_norm": 0.6758906245231628,
      "learning_rate": 3.0367590585362564e-06,
      "loss": 0.1313,
      "num_input_tokens_seen": 54130808,
      "step": 33660
    },
    {
      "epoch": 1.5240046175784876,
      "grad_norm": 0.7253269553184509,
      "learning_rate": 3.0320710627243813e-06,
      "loss": 0.1389,
      "num_input_tokens_seen": 54138648,
      "step": 33665
    },
    {
      "epoch": 1.5242309694651306,
      "grad_norm": 0.5907518267631531,
      "learning_rate": 3.027386454643222e-06,
      "loss": 0.1458,
      "num_input_tokens_seen": 54146392,
      "step": 33670
    },
    {
      "epoch": 1.5244573213517736,
      "grad_norm": 0.5920408368110657,
      "learning_rate": 3.0227052350151914e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 54154296,
      "step": 33675
    },
    {
      "epoch": 1.5246836732384166,
      "grad_norm": 0.8465604186058044,
      "learning_rate": 3.0180274045621957e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 54162168,
      "step": 33680
    },
    {
      "epoch": 1.5249100251250596,
      "grad_norm": 0.6137626767158508,
      "learning_rate": 3.013352964005625e-06,
      "loss": 0.1246,
      "num_input_tokens_seen": 54170648,
      "step": 33685
    },
    {
      "epoch": 1.5251363770117023,
      "grad_norm": 0.4683593809604645,
      "learning_rate": 3.0086819140663218e-06,
      "loss": 0.1362,
      "num_input_tokens_seen": 54178648,
      "step": 33690
    },
    {
      "epoch": 1.5253627288983453,
      "grad_norm": 0.6052393317222595,
      "learning_rate": 3.0040142554646265e-06,
      "loss": 0.1317,
      "num_input_tokens_seen": 54186616,
      "step": 33695
    },
    {
      "epoch": 1.5255890807849883,
      "grad_norm": 0.6991689801216125,
      "learning_rate": 2.999349988920361e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 54194936,
      "step": 33700
    },
    {
      "epoch": 1.5258154326716313,
      "grad_norm": 0.5682336091995239,
      "learning_rate": 2.994689115152796e-06,
      "loss": 0.1507,
      "num_input_tokens_seen": 54203544,
      "step": 33705
    },
    {
      "epoch": 1.5260417845582743,
      "grad_norm": 0.45137321949005127,
      "learning_rate": 2.9900316348807105e-06,
      "loss": 0.1414,
      "num_input_tokens_seen": 54211640,
      "step": 33710
    },
    {
      "epoch": 1.5262681364449173,
      "grad_norm": 1.1254955530166626,
      "learning_rate": 2.985377548822338e-06,
      "loss": 0.1318,
      "num_input_tokens_seen": 54219416,
      "step": 33715
    },
    {
      "epoch": 1.52649448833156,
      "grad_norm": 0.5029290318489075,
      "learning_rate": 2.980726857695404e-06,
      "loss": 0.123,
      "num_input_tokens_seen": 54227256,
      "step": 33720
    },
    {
      "epoch": 1.526720840218203,
      "grad_norm": 0.8196085095405579,
      "learning_rate": 2.9760795622171017e-06,
      "loss": 0.1345,
      "num_input_tokens_seen": 54234968,
      "step": 33725
    },
    {
      "epoch": 1.526947192104846,
      "grad_norm": 0.4516723155975342,
      "learning_rate": 2.971435663104094e-06,
      "loss": 0.1549,
      "num_input_tokens_seen": 54243192,
      "step": 33730
    },
    {
      "epoch": 1.527173543991489,
      "grad_norm": 0.574869692325592,
      "learning_rate": 2.9667951610725385e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 54251192,
      "step": 33735
    },
    {
      "epoch": 1.527399895878132,
      "grad_norm": 0.7901363968849182,
      "learning_rate": 2.9621580568380575e-06,
      "loss": 0.1793,
      "num_input_tokens_seen": 54259512,
      "step": 33740
    },
    {
      "epoch": 1.527626247764775,
      "grad_norm": 0.8827589154243469,
      "learning_rate": 2.9575243511157453e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 54267512,
      "step": 33745
    },
    {
      "epoch": 1.527852599651418,
      "grad_norm": 0.48363208770751953,
      "learning_rate": 2.952894044620186e-06,
      "loss": 0.1253,
      "num_input_tokens_seen": 54276120,
      "step": 33750
    },
    {
      "epoch": 1.528078951538061,
      "grad_norm": 0.353374719619751,
      "learning_rate": 2.948267138065419e-06,
      "loss": 0.1195,
      "num_input_tokens_seen": 54284248,
      "step": 33755
    },
    {
      "epoch": 1.528305303424704,
      "grad_norm": 0.5465307831764221,
      "learning_rate": 2.943643632164983e-06,
      "loss": 0.1204,
      "num_input_tokens_seen": 54292216,
      "step": 33760
    },
    {
      "epoch": 1.528531655311347,
      "grad_norm": 0.5666021108627319,
      "learning_rate": 2.939023527631879e-06,
      "loss": 0.1526,
      "num_input_tokens_seen": 54300440,
      "step": 33765
    },
    {
      "epoch": 1.52875800719799,
      "grad_norm": 1.1327857971191406,
      "learning_rate": 2.934406825178576e-06,
      "loss": 0.1478,
      "num_input_tokens_seen": 54308952,
      "step": 33770
    },
    {
      "epoch": 1.528984359084633,
      "grad_norm": 0.31310275197029114,
      "learning_rate": 2.9297935255170357e-06,
      "loss": 0.1322,
      "num_input_tokens_seen": 54317304,
      "step": 33775
    },
    {
      "epoch": 1.529210710971276,
      "grad_norm": 0.6270111799240112,
      "learning_rate": 2.925183629358691e-06,
      "loss": 0.1635,
      "num_input_tokens_seen": 54326168,
      "step": 33780
    },
    {
      "epoch": 1.529437062857919,
      "grad_norm": 0.37624475359916687,
      "learning_rate": 2.9205771374144346e-06,
      "loss": 0.1401,
      "num_input_tokens_seen": 54334520,
      "step": 33785
    },
    {
      "epoch": 1.529663414744562,
      "grad_norm": 0.7086783051490784,
      "learning_rate": 2.915974050394657e-06,
      "loss": 0.1179,
      "num_input_tokens_seen": 54342936,
      "step": 33790
    },
    {
      "epoch": 1.529889766631205,
      "grad_norm": 0.6464519500732422,
      "learning_rate": 2.9113743690092067e-06,
      "loss": 0.195,
      "num_input_tokens_seen": 54351480,
      "step": 33795
    },
    {
      "epoch": 1.530116118517848,
      "grad_norm": 0.6630606651306152,
      "learning_rate": 2.906778093967402e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 54359832,
      "step": 33800
    },
    {
      "epoch": 1.530116118517848,
      "eval_loss": 0.1437305361032486,
      "eval_runtime": 405.0856,
      "eval_samples_per_second": 96.945,
      "eval_steps_per_second": 24.237,
      "num_input_tokens_seen": 54359832,
      "step": 33800
    },
    {
      "epoch": 1.5303424704044908,
      "grad_norm": 0.5511282086372375,
      "learning_rate": 2.9021852259780656e-06,
      "loss": 0.1317,
      "num_input_tokens_seen": 54368536,
      "step": 33805
    },
    {
      "epoch": 1.5305688222911338,
      "grad_norm": 0.3048146367073059,
      "learning_rate": 2.8975957657494583e-06,
      "loss": 0.1545,
      "num_input_tokens_seen": 54376376,
      "step": 33810
    },
    {
      "epoch": 1.5307951741777768,
      "grad_norm": 0.7299084663391113,
      "learning_rate": 2.8930097139893417e-06,
      "loss": 0.1189,
      "num_input_tokens_seen": 54384664,
      "step": 33815
    },
    {
      "epoch": 1.5310215260644198,
      "grad_norm": 0.6065192222595215,
      "learning_rate": 2.888427071404945e-06,
      "loss": 0.169,
      "num_input_tokens_seen": 54393464,
      "step": 33820
    },
    {
      "epoch": 1.5312478779510628,
      "grad_norm": 0.7954663038253784,
      "learning_rate": 2.8838478387029606e-06,
      "loss": 0.1369,
      "num_input_tokens_seen": 54400824,
      "step": 33825
    },
    {
      "epoch": 1.5314742298377055,
      "grad_norm": 0.6469586491584778,
      "learning_rate": 2.8792720165895737e-06,
      "loss": 0.1563,
      "num_input_tokens_seen": 54409208,
      "step": 33830
    },
    {
      "epoch": 1.5317005817243485,
      "grad_norm": 0.3353521525859833,
      "learning_rate": 2.874699605770423e-06,
      "loss": 0.1494,
      "num_input_tokens_seen": 54417208,
      "step": 33835
    },
    {
      "epoch": 1.5319269336109915,
      "grad_norm": 0.441864937543869,
      "learning_rate": 2.8701306069506383e-06,
      "loss": 0.1034,
      "num_input_tokens_seen": 54425208,
      "step": 33840
    },
    {
      "epoch": 1.5321532854976345,
      "grad_norm": 0.39031484723091125,
      "learning_rate": 2.8655650208348178e-06,
      "loss": 0.1361,
      "num_input_tokens_seen": 54432984,
      "step": 33845
    },
    {
      "epoch": 1.5323796373842775,
      "grad_norm": 0.4480418562889099,
      "learning_rate": 2.8610028481270257e-06,
      "loss": 0.1268,
      "num_input_tokens_seen": 54441144,
      "step": 33850
    },
    {
      "epoch": 1.5326059892709205,
      "grad_norm": 0.7208046913146973,
      "learning_rate": 2.856444089530813e-06,
      "loss": 0.1107,
      "num_input_tokens_seen": 54448632,
      "step": 33855
    },
    {
      "epoch": 1.5328323411575635,
      "grad_norm": 0.5006346106529236,
      "learning_rate": 2.8518887457491955e-06,
      "loss": 0.1195,
      "num_input_tokens_seen": 54456664,
      "step": 33860
    },
    {
      "epoch": 1.5330586930442065,
      "grad_norm": 0.4714199900627136,
      "learning_rate": 2.8473368174846666e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 54464376,
      "step": 33865
    },
    {
      "epoch": 1.5332850449308495,
      "grad_norm": 1.3758668899536133,
      "learning_rate": 2.842788305439184e-06,
      "loss": 0.1525,
      "num_input_tokens_seen": 54472600,
      "step": 33870
    },
    {
      "epoch": 1.5335113968174925,
      "grad_norm": 0.49429723620414734,
      "learning_rate": 2.8382432103141925e-06,
      "loss": 0.1405,
      "num_input_tokens_seen": 54480440,
      "step": 33875
    },
    {
      "epoch": 1.5337377487041355,
      "grad_norm": 0.5011327862739563,
      "learning_rate": 2.833701532810598e-06,
      "loss": 0.1408,
      "num_input_tokens_seen": 54488952,
      "step": 33880
    },
    {
      "epoch": 1.5339641005907785,
      "grad_norm": 0.5185560584068298,
      "learning_rate": 2.8291632736287877e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 54496760,
      "step": 33885
    },
    {
      "epoch": 1.5341904524774215,
      "grad_norm": 0.3541467487812042,
      "learning_rate": 2.824628433468615e-06,
      "loss": 0.1226,
      "num_input_tokens_seen": 54504120,
      "step": 33890
    },
    {
      "epoch": 1.5344168043640645,
      "grad_norm": 0.5039882659912109,
      "learning_rate": 2.8200970130294073e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 54511736,
      "step": 33895
    },
    {
      "epoch": 1.5346431562507075,
      "grad_norm": 0.6853495836257935,
      "learning_rate": 2.8155690130099775e-06,
      "loss": 0.1003,
      "num_input_tokens_seen": 54519512,
      "step": 33900
    },
    {
      "epoch": 1.5348695081373505,
      "grad_norm": 0.5029970407485962,
      "learning_rate": 2.8110444341085895e-06,
      "loss": 0.1452,
      "num_input_tokens_seen": 54528248,
      "step": 33905
    },
    {
      "epoch": 1.5350958600239935,
      "grad_norm": 0.9384796023368835,
      "learning_rate": 2.806523277022996e-06,
      "loss": 0.1387,
      "num_input_tokens_seen": 54536536,
      "step": 33910
    },
    {
      "epoch": 1.5353222119106362,
      "grad_norm": 0.8784890174865723,
      "learning_rate": 2.802005542450409e-06,
      "loss": 0.121,
      "num_input_tokens_seen": 54544312,
      "step": 33915
    },
    {
      "epoch": 1.5355485637972792,
      "grad_norm": 1.029774785041809,
      "learning_rate": 2.797491231087526e-06,
      "loss": 0.1301,
      "num_input_tokens_seen": 54552408,
      "step": 33920
    },
    {
      "epoch": 1.5357749156839222,
      "grad_norm": 0.42809122800827026,
      "learning_rate": 2.7929803436305137e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 54560472,
      "step": 33925
    },
    {
      "epoch": 1.5360012675705652,
      "grad_norm": 0.7107066512107849,
      "learning_rate": 2.788472880774998e-06,
      "loss": 0.1472,
      "num_input_tokens_seen": 54567736,
      "step": 33930
    },
    {
      "epoch": 1.5362276194572082,
      "grad_norm": 0.7064847350120544,
      "learning_rate": 2.7839688432160977e-06,
      "loss": 0.1212,
      "num_input_tokens_seen": 54575992,
      "step": 33935
    },
    {
      "epoch": 1.5364539713438512,
      "grad_norm": 0.6588173508644104,
      "learning_rate": 2.779468231648383e-06,
      "loss": 0.1329,
      "num_input_tokens_seen": 54584376,
      "step": 33940
    },
    {
      "epoch": 1.536680323230494,
      "grad_norm": 0.6047943234443665,
      "learning_rate": 2.774971046765906e-06,
      "loss": 0.1015,
      "num_input_tokens_seen": 54592280,
      "step": 33945
    },
    {
      "epoch": 1.536906675117137,
      "grad_norm": 0.6930447220802307,
      "learning_rate": 2.770477289262194e-06,
      "loss": 0.141,
      "num_input_tokens_seen": 54600472,
      "step": 33950
    },
    {
      "epoch": 1.53713302700378,
      "grad_norm": 0.3330308496952057,
      "learning_rate": 2.765986959830233e-06,
      "loss": 0.1796,
      "num_input_tokens_seen": 54608152,
      "step": 33955
    },
    {
      "epoch": 1.537359378890423,
      "grad_norm": 1.0782667398452759,
      "learning_rate": 2.761500059162492e-06,
      "loss": 0.1489,
      "num_input_tokens_seen": 54616248,
      "step": 33960
    },
    {
      "epoch": 1.537585730777066,
      "grad_norm": 0.7216376662254333,
      "learning_rate": 2.757016587950914e-06,
      "loss": 0.1228,
      "num_input_tokens_seen": 54624376,
      "step": 33965
    },
    {
      "epoch": 1.537812082663709,
      "grad_norm": 1.8870537281036377,
      "learning_rate": 2.752536546886897e-06,
      "loss": 0.1512,
      "num_input_tokens_seen": 54632312,
      "step": 33970
    },
    {
      "epoch": 1.538038434550352,
      "grad_norm": 1.2740753889083862,
      "learning_rate": 2.7480599366613234e-06,
      "loss": 0.1141,
      "num_input_tokens_seen": 54640536,
      "step": 33975
    },
    {
      "epoch": 1.538264786436995,
      "grad_norm": 0.6587880849838257,
      "learning_rate": 2.7435867579645473e-06,
      "loss": 0.1871,
      "num_input_tokens_seen": 54648312,
      "step": 33980
    },
    {
      "epoch": 1.538491138323638,
      "grad_norm": 0.6821882128715515,
      "learning_rate": 2.739117011486378e-06,
      "loss": 0.1349,
      "num_input_tokens_seen": 54656184,
      "step": 33985
    },
    {
      "epoch": 1.538717490210281,
      "grad_norm": 0.452199786901474,
      "learning_rate": 2.7346506979161216e-06,
      "loss": 0.1697,
      "num_input_tokens_seen": 54664440,
      "step": 33990
    },
    {
      "epoch": 1.538943842096924,
      "grad_norm": 0.44006308913230896,
      "learning_rate": 2.7301878179425227e-06,
      "loss": 0.1169,
      "num_input_tokens_seen": 54672184,
      "step": 33995
    },
    {
      "epoch": 1.539170193983567,
      "grad_norm": 0.8609959483146667,
      "learning_rate": 2.7257283722538244e-06,
      "loss": 0.1076,
      "num_input_tokens_seen": 54681080,
      "step": 34000
    },
    {
      "epoch": 1.539170193983567,
      "eval_loss": 0.14348752796649933,
      "eval_runtime": 404.8405,
      "eval_samples_per_second": 97.004,
      "eval_steps_per_second": 24.252,
      "num_input_tokens_seen": 54681080,
      "step": 34000
    },
    {
      "epoch": 1.53939654587021,
      "grad_norm": 1.347143530845642,
      "learning_rate": 2.7212723615377326e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 54688856,
      "step": 34005
    },
    {
      "epoch": 1.539622897756853,
      "grad_norm": 0.8640576601028442,
      "learning_rate": 2.7168197864814145e-06,
      "loss": 0.1269,
      "num_input_tokens_seen": 54696600,
      "step": 34010
    },
    {
      "epoch": 1.539849249643496,
      "grad_norm": 0.341683030128479,
      "learning_rate": 2.712370647771509e-06,
      "loss": 0.1196,
      "num_input_tokens_seen": 54704440,
      "step": 34015
    },
    {
      "epoch": 1.540075601530139,
      "grad_norm": 0.9373278021812439,
      "learning_rate": 2.707924946094137e-06,
      "loss": 0.134,
      "num_input_tokens_seen": 54713048,
      "step": 34020
    },
    {
      "epoch": 1.5403019534167819,
      "grad_norm": 1.2403981685638428,
      "learning_rate": 2.7034826821348723e-06,
      "loss": 0.1467,
      "num_input_tokens_seen": 54721176,
      "step": 34025
    },
    {
      "epoch": 1.5405283053034247,
      "grad_norm": 0.4894223213195801,
      "learning_rate": 2.6990438565787786e-06,
      "loss": 0.1692,
      "num_input_tokens_seen": 54729368,
      "step": 34030
    },
    {
      "epoch": 1.5407546571900677,
      "grad_norm": 0.4864262044429779,
      "learning_rate": 2.6946084701103714e-06,
      "loss": 0.137,
      "num_input_tokens_seen": 54737080,
      "step": 34035
    },
    {
      "epoch": 1.5409810090767107,
      "grad_norm": 0.49452289938926697,
      "learning_rate": 2.6901765234136428e-06,
      "loss": 0.1821,
      "num_input_tokens_seen": 54745752,
      "step": 34040
    },
    {
      "epoch": 1.5412073609633536,
      "grad_norm": 0.394873708486557,
      "learning_rate": 2.685748017172063e-06,
      "loss": 0.1463,
      "num_input_tokens_seen": 54753976,
      "step": 34045
    },
    {
      "epoch": 1.5414337128499966,
      "grad_norm": 1.0214654207229614,
      "learning_rate": 2.681322952068549e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 54762456,
      "step": 34050
    },
    {
      "epoch": 1.5416600647366396,
      "grad_norm": 0.5624420046806335,
      "learning_rate": 2.6769013287855137e-06,
      "loss": 0.1457,
      "num_input_tokens_seen": 54770712,
      "step": 34055
    },
    {
      "epoch": 1.5418864166232824,
      "grad_norm": 0.9620119333267212,
      "learning_rate": 2.6724831480048286e-06,
      "loss": 0.1593,
      "num_input_tokens_seen": 54778616,
      "step": 34060
    },
    {
      "epoch": 1.5421127685099254,
      "grad_norm": 0.6415120959281921,
      "learning_rate": 2.66806841040782e-06,
      "loss": 0.1227,
      "num_input_tokens_seen": 54786136,
      "step": 34065
    },
    {
      "epoch": 1.5423391203965684,
      "grad_norm": 0.9270049929618835,
      "learning_rate": 2.6636571166753083e-06,
      "loss": 0.171,
      "num_input_tokens_seen": 54794360,
      "step": 34070
    },
    {
      "epoch": 1.5425654722832114,
      "grad_norm": 0.39248406887054443,
      "learning_rate": 2.6592492674875598e-06,
      "loss": 0.1265,
      "num_input_tokens_seen": 54802424,
      "step": 34075
    },
    {
      "epoch": 1.5427918241698544,
      "grad_norm": 0.2483447790145874,
      "learning_rate": 2.6548448635243305e-06,
      "loss": 0.108,
      "num_input_tokens_seen": 54810840,
      "step": 34080
    },
    {
      "epoch": 1.5430181760564974,
      "grad_norm": 0.9916409850120544,
      "learning_rate": 2.650443905464828e-06,
      "loss": 0.1211,
      "num_input_tokens_seen": 54818776,
      "step": 34085
    },
    {
      "epoch": 1.5432445279431404,
      "grad_norm": 0.3802112638950348,
      "learning_rate": 2.646046393987739e-06,
      "loss": 0.1278,
      "num_input_tokens_seen": 54827288,
      "step": 34090
    },
    {
      "epoch": 1.5434708798297834,
      "grad_norm": 1.069146752357483,
      "learning_rate": 2.64165232977121e-06,
      "loss": 0.1655,
      "num_input_tokens_seen": 54835768,
      "step": 34095
    },
    {
      "epoch": 1.5436972317164264,
      "grad_norm": 0.8970841765403748,
      "learning_rate": 2.6372617134928695e-06,
      "loss": 0.1309,
      "num_input_tokens_seen": 54843288,
      "step": 34100
    },
    {
      "epoch": 1.5439235836030694,
      "grad_norm": 1.0662099123001099,
      "learning_rate": 2.6328745458297943e-06,
      "loss": 0.1302,
      "num_input_tokens_seen": 54850936,
      "step": 34105
    },
    {
      "epoch": 1.5441499354897124,
      "grad_norm": 0.4317435622215271,
      "learning_rate": 2.6284908274585546e-06,
      "loss": 0.1702,
      "num_input_tokens_seen": 54859032,
      "step": 34110
    },
    {
      "epoch": 1.5443762873763554,
      "grad_norm": 0.3714941442012787,
      "learning_rate": 2.6241105590551595e-06,
      "loss": 0.1112,
      "num_input_tokens_seen": 54867256,
      "step": 34115
    },
    {
      "epoch": 1.5446026392629983,
      "grad_norm": 0.5984408259391785,
      "learning_rate": 2.6197337412951105e-06,
      "loss": 0.1159,
      "num_input_tokens_seen": 54874744,
      "step": 34120
    },
    {
      "epoch": 1.5448289911496413,
      "grad_norm": 0.635013222694397,
      "learning_rate": 2.6153603748533705e-06,
      "loss": 0.1524,
      "num_input_tokens_seen": 54882648,
      "step": 34125
    },
    {
      "epoch": 1.5450553430362843,
      "grad_norm": 1.2134813070297241,
      "learning_rate": 2.6109904604043585e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 54890200,
      "step": 34130
    },
    {
      "epoch": 1.5452816949229273,
      "grad_norm": 0.7750403881072998,
      "learning_rate": 2.6066239986219765e-06,
      "loss": 0.1586,
      "num_input_tokens_seen": 54897624,
      "step": 34135
    },
    {
      "epoch": 1.5455080468095703,
      "grad_norm": 0.5110148787498474,
      "learning_rate": 2.602260990179592e-06,
      "loss": 0.1568,
      "num_input_tokens_seen": 54905720,
      "step": 34140
    },
    {
      "epoch": 1.545734398696213,
      "grad_norm": 0.3119056224822998,
      "learning_rate": 2.5979014357500248e-06,
      "loss": 0.1295,
      "num_input_tokens_seen": 54913880,
      "step": 34145
    },
    {
      "epoch": 1.545960750582856,
      "grad_norm": 0.5358433127403259,
      "learning_rate": 2.5935453360055844e-06,
      "loss": 0.1598,
      "num_input_tokens_seen": 54921720,
      "step": 34150
    },
    {
      "epoch": 1.546187102469499,
      "grad_norm": 0.6766946315765381,
      "learning_rate": 2.5891926916180283e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 54930072,
      "step": 34155
    },
    {
      "epoch": 1.546413454356142,
      "grad_norm": 0.4497126340866089,
      "learning_rate": 2.5848435032585883e-06,
      "loss": 0.1592,
      "num_input_tokens_seen": 54938328,
      "step": 34160
    },
    {
      "epoch": 1.546639806242785,
      "grad_norm": 0.31789013743400574,
      "learning_rate": 2.58049777159797e-06,
      "loss": 0.1377,
      "num_input_tokens_seen": 54946200,
      "step": 34165
    },
    {
      "epoch": 1.5468661581294278,
      "grad_norm": 0.7994097471237183,
      "learning_rate": 2.576155497306332e-06,
      "loss": 0.1599,
      "num_input_tokens_seen": 54955160,
      "step": 34170
    },
    {
      "epoch": 1.5470925100160708,
      "grad_norm": 0.8937859535217285,
      "learning_rate": 2.57181668105331e-06,
      "loss": 0.149,
      "num_input_tokens_seen": 54963352,
      "step": 34175
    },
    {
      "epoch": 1.5473188619027138,
      "grad_norm": 1.116839051246643,
      "learning_rate": 2.567481323508014e-06,
      "loss": 0.1197,
      "num_input_tokens_seen": 54971704,
      "step": 34180
    },
    {
      "epoch": 1.5475452137893568,
      "grad_norm": 0.47417670488357544,
      "learning_rate": 2.5631494253389954e-06,
      "loss": 0.1822,
      "num_input_tokens_seen": 54980600,
      "step": 34185
    },
    {
      "epoch": 1.5477715656759998,
      "grad_norm": 0.37651461362838745,
      "learning_rate": 2.5588209872142997e-06,
      "loss": 0.1233,
      "num_input_tokens_seen": 54988248,
      "step": 34190
    },
    {
      "epoch": 1.5479979175626428,
      "grad_norm": 0.876819372177124,
      "learning_rate": 2.5544960098014186e-06,
      "loss": 0.1509,
      "num_input_tokens_seen": 54996824,
      "step": 34195
    },
    {
      "epoch": 1.5482242694492858,
      "grad_norm": 0.5660488605499268,
      "learning_rate": 2.550174493767318e-06,
      "loss": 0.0996,
      "num_input_tokens_seen": 55004920,
      "step": 34200
    },
    {
      "epoch": 1.5482242694492858,
      "eval_loss": 0.14316494762897491,
      "eval_runtime": 404.0125,
      "eval_samples_per_second": 97.202,
      "eval_steps_per_second": 24.301,
      "num_input_tokens_seen": 55004920,
      "step": 34200
    },
    {
      "epoch": 1.5484506213359288,
      "grad_norm": 0.85444575548172,
      "learning_rate": 2.545856439778438e-06,
      "loss": 0.153,
      "num_input_tokens_seen": 55013432,
      "step": 34205
    },
    {
      "epoch": 1.5486769732225718,
      "grad_norm": 0.46959537267684937,
      "learning_rate": 2.541541848500667e-06,
      "loss": 0.1577,
      "num_input_tokens_seen": 55021816,
      "step": 34210
    },
    {
      "epoch": 1.5489033251092148,
      "grad_norm": 0.24393613636493683,
      "learning_rate": 2.5372307205993733e-06,
      "loss": 0.1243,
      "num_input_tokens_seen": 55029592,
      "step": 34215
    },
    {
      "epoch": 1.5491296769958578,
      "grad_norm": 0.49366796016693115,
      "learning_rate": 2.5329230567393917e-06,
      "loss": 0.1069,
      "num_input_tokens_seen": 55037464,
      "step": 34220
    },
    {
      "epoch": 1.5493560288825008,
      "grad_norm": 0.49206772446632385,
      "learning_rate": 2.5286188575850164e-06,
      "loss": 0.1422,
      "num_input_tokens_seen": 55045784,
      "step": 34225
    },
    {
      "epoch": 1.5495823807691438,
      "grad_norm": 0.5754907131195068,
      "learning_rate": 2.5243181237999984e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 55054104,
      "step": 34230
    },
    {
      "epoch": 1.5498087326557868,
      "grad_norm": 0.905153214931488,
      "learning_rate": 2.520020856047578e-06,
      "loss": 0.1596,
      "num_input_tokens_seen": 55062552,
      "step": 34235
    },
    {
      "epoch": 1.5500350845424298,
      "grad_norm": 0.4099457263946533,
      "learning_rate": 2.515727054990438e-06,
      "loss": 0.16,
      "num_input_tokens_seen": 55070264,
      "step": 34240
    },
    {
      "epoch": 1.5502614364290728,
      "grad_norm": 0.6191556453704834,
      "learning_rate": 2.511436721290747e-06,
      "loss": 0.1088,
      "num_input_tokens_seen": 55078072,
      "step": 34245
    },
    {
      "epoch": 1.5504877883157158,
      "grad_norm": 0.8223111629486084,
      "learning_rate": 2.5071498556101164e-06,
      "loss": 0.1318,
      "num_input_tokens_seen": 55086296,
      "step": 34250
    },
    {
      "epoch": 1.5507141402023585,
      "grad_norm": 0.34987354278564453,
      "learning_rate": 2.5028664586096485e-06,
      "loss": 0.1021,
      "num_input_tokens_seen": 55093912,
      "step": 34255
    },
    {
      "epoch": 1.5509404920890015,
      "grad_norm": 0.4824505150318146,
      "learning_rate": 2.498586530949881e-06,
      "loss": 0.1288,
      "num_input_tokens_seen": 55102520,
      "step": 34260
    },
    {
      "epoch": 1.5511668439756445,
      "grad_norm": 0.7929821610450745,
      "learning_rate": 2.4943100732908427e-06,
      "loss": 0.1229,
      "num_input_tokens_seen": 55110488,
      "step": 34265
    },
    {
      "epoch": 1.5513931958622875,
      "grad_norm": 0.982363760471344,
      "learning_rate": 2.4900370862920188e-06,
      "loss": 0.1662,
      "num_input_tokens_seen": 55119192,
      "step": 34270
    },
    {
      "epoch": 1.5516195477489305,
      "grad_norm": 0.47214192152023315,
      "learning_rate": 2.4857675706123518e-06,
      "loss": 0.1393,
      "num_input_tokens_seen": 55127192,
      "step": 34275
    },
    {
      "epoch": 1.5518458996355735,
      "grad_norm": 0.9766258597373962,
      "learning_rate": 2.4815015269102543e-06,
      "loss": 0.1697,
      "num_input_tokens_seen": 55135128,
      "step": 34280
    },
    {
      "epoch": 1.5520722515222163,
      "grad_norm": 0.9432665705680847,
      "learning_rate": 2.477238955843611e-06,
      "loss": 0.1443,
      "num_input_tokens_seen": 55142776,
      "step": 34285
    },
    {
      "epoch": 1.5522986034088593,
      "grad_norm": 0.3812001645565033,
      "learning_rate": 2.4729798580697573e-06,
      "loss": 0.1748,
      "num_input_tokens_seen": 55150520,
      "step": 34290
    },
    {
      "epoch": 1.5525249552955023,
      "grad_norm": 0.38554808497428894,
      "learning_rate": 2.4687242342455034e-06,
      "loss": 0.1428,
      "num_input_tokens_seen": 55158104,
      "step": 34295
    },
    {
      "epoch": 1.5527513071821453,
      "grad_norm": 0.8783165216445923,
      "learning_rate": 2.4644720850271196e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 55166264,
      "step": 34300
    },
    {
      "epoch": 1.5529776590687883,
      "grad_norm": 0.5250450968742371,
      "learning_rate": 2.4602234110703364e-06,
      "loss": 0.1078,
      "num_input_tokens_seen": 55174488,
      "step": 34305
    },
    {
      "epoch": 1.5532040109554313,
      "grad_norm": 0.4601297378540039,
      "learning_rate": 2.4559782130303576e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 55182040,
      "step": 34310
    },
    {
      "epoch": 1.5534303628420743,
      "grad_norm": 0.764395534992218,
      "learning_rate": 2.451736491561843e-06,
      "loss": 0.2039,
      "num_input_tokens_seen": 55190456,
      "step": 34315
    },
    {
      "epoch": 1.5536567147287172,
      "grad_norm": 0.33915603160858154,
      "learning_rate": 2.4474982473189163e-06,
      "loss": 0.1227,
      "num_input_tokens_seen": 55198616,
      "step": 34320
    },
    {
      "epoch": 1.5538830666153602,
      "grad_norm": 0.7540341019630432,
      "learning_rate": 2.4432634809551796e-06,
      "loss": 0.1203,
      "num_input_tokens_seen": 55206808,
      "step": 34325
    },
    {
      "epoch": 1.5541094185020032,
      "grad_norm": 1.1014102697372437,
      "learning_rate": 2.439032193123675e-06,
      "loss": 0.1099,
      "num_input_tokens_seen": 55214200,
      "step": 34330
    },
    {
      "epoch": 1.5543357703886462,
      "grad_norm": 0.8861912488937378,
      "learning_rate": 2.4348043844769297e-06,
      "loss": 0.1923,
      "num_input_tokens_seen": 55222168,
      "step": 34335
    },
    {
      "epoch": 1.5545621222752892,
      "grad_norm": 0.6523618698120117,
      "learning_rate": 2.4305800556669146e-06,
      "loss": 0.1915,
      "num_input_tokens_seen": 55230008,
      "step": 34340
    },
    {
      "epoch": 1.5547884741619322,
      "grad_norm": 0.5689152479171753,
      "learning_rate": 2.426359207345083e-06,
      "loss": 0.1169,
      "num_input_tokens_seen": 55238040,
      "step": 34345
    },
    {
      "epoch": 1.5550148260485752,
      "grad_norm": 0.5014533996582031,
      "learning_rate": 2.4221418401623396e-06,
      "loss": 0.1386,
      "num_input_tokens_seen": 55245944,
      "step": 34350
    },
    {
      "epoch": 1.5552411779352182,
      "grad_norm": 0.4600675404071808,
      "learning_rate": 2.4179279547690557e-06,
      "loss": 0.1269,
      "num_input_tokens_seen": 55254392,
      "step": 34355
    },
    {
      "epoch": 1.5554675298218612,
      "grad_norm": 0.8950806856155396,
      "learning_rate": 2.413717551815062e-06,
      "loss": 0.1689,
      "num_input_tokens_seen": 55262456,
      "step": 34360
    },
    {
      "epoch": 1.5556938817085042,
      "grad_norm": 0.906383216381073,
      "learning_rate": 2.409510631949666e-06,
      "loss": 0.1626,
      "num_input_tokens_seen": 55270328,
      "step": 34365
    },
    {
      "epoch": 1.555920233595147,
      "grad_norm": 0.662895143032074,
      "learning_rate": 2.405307195821618e-06,
      "loss": 0.1089,
      "num_input_tokens_seen": 55278360,
      "step": 34370
    },
    {
      "epoch": 1.55614658548179,
      "grad_norm": 0.3762154281139374,
      "learning_rate": 2.4011072440791372e-06,
      "loss": 0.1536,
      "num_input_tokens_seen": 55286232,
      "step": 34375
    },
    {
      "epoch": 1.556372937368433,
      "grad_norm": 0.6333240270614624,
      "learning_rate": 2.3969107773699233e-06,
      "loss": 0.166,
      "num_input_tokens_seen": 55294392,
      "step": 34380
    },
    {
      "epoch": 1.556599289255076,
      "grad_norm": 0.6228295564651489,
      "learning_rate": 2.3927177963411096e-06,
      "loss": 0.1851,
      "num_input_tokens_seen": 55302936,
      "step": 34385
    },
    {
      "epoch": 1.556825641141719,
      "grad_norm": 0.4112420678138733,
      "learning_rate": 2.3885283016393144e-06,
      "loss": 0.1675,
      "num_input_tokens_seen": 55311064,
      "step": 34390
    },
    {
      "epoch": 1.5570519930283617,
      "grad_norm": 0.9015906453132629,
      "learning_rate": 2.3843422939106076e-06,
      "loss": 0.1421,
      "num_input_tokens_seen": 55319320,
      "step": 34395
    },
    {
      "epoch": 1.5572783449150047,
      "grad_norm": 0.849987804889679,
      "learning_rate": 2.380159773800525e-06,
      "loss": 0.1136,
      "num_input_tokens_seen": 55328216,
      "step": 34400
    },
    {
      "epoch": 1.5572783449150047,
      "eval_loss": 0.1433349847793579,
      "eval_runtime": 405.4345,
      "eval_samples_per_second": 96.862,
      "eval_steps_per_second": 24.216,
      "num_input_tokens_seen": 55328216,
      "step": 34400
    },
    {
      "epoch": 1.5575046968016477,
      "grad_norm": 0.42490828037261963,
      "learning_rate": 2.3759807419540675e-06,
      "loss": 0.1564,
      "num_input_tokens_seen": 55335704,
      "step": 34405
    },
    {
      "epoch": 1.5577310486882907,
      "grad_norm": 0.9637981653213501,
      "learning_rate": 2.3718051990156835e-06,
      "loss": 0.1819,
      "num_input_tokens_seen": 55343512,
      "step": 34410
    },
    {
      "epoch": 1.5579574005749337,
      "grad_norm": 0.3710101246833801,
      "learning_rate": 2.367633145629311e-06,
      "loss": 0.1548,
      "num_input_tokens_seen": 55351800,
      "step": 34415
    },
    {
      "epoch": 1.5581837524615767,
      "grad_norm": 0.2973514199256897,
      "learning_rate": 2.363464582438316e-06,
      "loss": 0.1274,
      "num_input_tokens_seen": 55360248,
      "step": 34420
    },
    {
      "epoch": 1.5584101043482197,
      "grad_norm": 0.3485260307788849,
      "learning_rate": 2.3592995100855526e-06,
      "loss": 0.0981,
      "num_input_tokens_seen": 55368152,
      "step": 34425
    },
    {
      "epoch": 1.5586364562348627,
      "grad_norm": 0.816257655620575,
      "learning_rate": 2.3551379292133273e-06,
      "loss": 0.169,
      "num_input_tokens_seen": 55376440,
      "step": 34430
    },
    {
      "epoch": 1.5588628081215057,
      "grad_norm": 0.57667076587677,
      "learning_rate": 2.3509798404634047e-06,
      "loss": 0.1478,
      "num_input_tokens_seen": 55384536,
      "step": 34435
    },
    {
      "epoch": 1.5590891600081487,
      "grad_norm": 1.0568667650222778,
      "learning_rate": 2.346825244477019e-06,
      "loss": 0.1223,
      "num_input_tokens_seen": 55393400,
      "step": 34440
    },
    {
      "epoch": 1.5593155118947917,
      "grad_norm": 0.681519091129303,
      "learning_rate": 2.3426741418948545e-06,
      "loss": 0.1544,
      "num_input_tokens_seen": 55401304,
      "step": 34445
    },
    {
      "epoch": 1.5595418637814347,
      "grad_norm": 0.7661498785018921,
      "learning_rate": 2.3385265333570715e-06,
      "loss": 0.1543,
      "num_input_tokens_seen": 55409688,
      "step": 34450
    },
    {
      "epoch": 1.5597682156680777,
      "grad_norm": 0.5054670572280884,
      "learning_rate": 2.334382419503278e-06,
      "loss": 0.1505,
      "num_input_tokens_seen": 55418040,
      "step": 34455
    },
    {
      "epoch": 1.5599945675547207,
      "grad_norm": 0.3500995337963104,
      "learning_rate": 2.3302418009725465e-06,
      "loss": 0.1162,
      "num_input_tokens_seen": 55426264,
      "step": 34460
    },
    {
      "epoch": 1.5602209194413637,
      "grad_norm": 1.1805187463760376,
      "learning_rate": 2.326104678403415e-06,
      "loss": 0.1365,
      "num_input_tokens_seen": 55434200,
      "step": 34465
    },
    {
      "epoch": 1.5604472713280066,
      "grad_norm": 0.5209856033325195,
      "learning_rate": 2.321971052433883e-06,
      "loss": 0.1464,
      "num_input_tokens_seen": 55442040,
      "step": 34470
    },
    {
      "epoch": 1.5606736232146496,
      "grad_norm": 0.6542225480079651,
      "learning_rate": 2.3178409237014004e-06,
      "loss": 0.1382,
      "num_input_tokens_seen": 55449752,
      "step": 34475
    },
    {
      "epoch": 1.5608999751012924,
      "grad_norm": 0.8985762596130371,
      "learning_rate": 2.313714292842889e-06,
      "loss": 0.139,
      "num_input_tokens_seen": 55457240,
      "step": 34480
    },
    {
      "epoch": 1.5611263269879354,
      "grad_norm": 0.4433375597000122,
      "learning_rate": 2.309591160494734e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 55465208,
      "step": 34485
    },
    {
      "epoch": 1.5613526788745784,
      "grad_norm": 0.36887967586517334,
      "learning_rate": 2.305471527292763e-06,
      "loss": 0.1609,
      "num_input_tokens_seen": 55473368,
      "step": 34490
    },
    {
      "epoch": 1.5615790307612214,
      "grad_norm": 0.47464460134506226,
      "learning_rate": 2.3013553938722817e-06,
      "loss": 0.1466,
      "num_input_tokens_seen": 55481816,
      "step": 34495
    },
    {
      "epoch": 1.5618053826478644,
      "grad_norm": 0.5438457131385803,
      "learning_rate": 2.297242760868043e-06,
      "loss": 0.1311,
      "num_input_tokens_seen": 55490520,
      "step": 34500
    },
    {
      "epoch": 1.5620317345345074,
      "grad_norm": 0.24879436194896698,
      "learning_rate": 2.2931336289142735e-06,
      "loss": 0.1185,
      "num_input_tokens_seen": 55498680,
      "step": 34505
    },
    {
      "epoch": 1.5622580864211502,
      "grad_norm": 0.5209116339683533,
      "learning_rate": 2.289027998644655e-06,
      "loss": 0.1201,
      "num_input_tokens_seen": 55506104,
      "step": 34510
    },
    {
      "epoch": 1.5624844383077932,
      "grad_norm": 0.8215847015380859,
      "learning_rate": 2.2849258706923228e-06,
      "loss": 0.1415,
      "num_input_tokens_seen": 55513624,
      "step": 34515
    },
    {
      "epoch": 1.5627107901944362,
      "grad_norm": 0.48212194442749023,
      "learning_rate": 2.2808272456898705e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 55521656,
      "step": 34520
    },
    {
      "epoch": 1.5629371420810791,
      "grad_norm": 0.645501434803009,
      "learning_rate": 2.2767321242693707e-06,
      "loss": 0.1494,
      "num_input_tokens_seen": 55529880,
      "step": 34525
    },
    {
      "epoch": 1.5631634939677221,
      "grad_norm": 0.43888333439826965,
      "learning_rate": 2.272640507062329e-06,
      "loss": 0.1345,
      "num_input_tokens_seen": 55537656,
      "step": 34530
    },
    {
      "epoch": 1.5633898458543651,
      "grad_norm": 1.187086820602417,
      "learning_rate": 2.2685523946997382e-06,
      "loss": 0.1145,
      "num_input_tokens_seen": 55545496,
      "step": 34535
    },
    {
      "epoch": 1.5636161977410081,
      "grad_norm": 0.7281479835510254,
      "learning_rate": 2.2644677878120245e-06,
      "loss": 0.1471,
      "num_input_tokens_seen": 55553880,
      "step": 34540
    },
    {
      "epoch": 1.5638425496276511,
      "grad_norm": 1.121429443359375,
      "learning_rate": 2.2603866870290897e-06,
      "loss": 0.1556,
      "num_input_tokens_seen": 55561304,
      "step": 34545
    },
    {
      "epoch": 1.5640689015142941,
      "grad_norm": 1.1390191316604614,
      "learning_rate": 2.256309092980294e-06,
      "loss": 0.1645,
      "num_input_tokens_seen": 55569112,
      "step": 34550
    },
    {
      "epoch": 1.5642952534009371,
      "grad_norm": 0.7268572449684143,
      "learning_rate": 2.252235006294448e-06,
      "loss": 0.1594,
      "num_input_tokens_seen": 55577496,
      "step": 34555
    },
    {
      "epoch": 1.56452160528758,
      "grad_norm": 0.6768231391906738,
      "learning_rate": 2.2481644275998333e-06,
      "loss": 0.1231,
      "num_input_tokens_seen": 55585624,
      "step": 34560
    },
    {
      "epoch": 1.564747957174223,
      "grad_norm": 0.7815474271774292,
      "learning_rate": 2.2440973575241832e-06,
      "loss": 0.1744,
      "num_input_tokens_seen": 55593816,
      "step": 34565
    },
    {
      "epoch": 1.564974309060866,
      "grad_norm": 0.6430264115333557,
      "learning_rate": 2.240033796694685e-06,
      "loss": 0.1871,
      "num_input_tokens_seen": 55602232,
      "step": 34570
    },
    {
      "epoch": 1.565200660947509,
      "grad_norm": 1.1708719730377197,
      "learning_rate": 2.235973745737999e-06,
      "loss": 0.1259,
      "num_input_tokens_seen": 55610488,
      "step": 34575
    },
    {
      "epoch": 1.565427012834152,
      "grad_norm": 0.695411741733551,
      "learning_rate": 2.2319172052802263e-06,
      "loss": 0.145,
      "num_input_tokens_seen": 55618616,
      "step": 34580
    },
    {
      "epoch": 1.565653364720795,
      "grad_norm": 0.49422410130500793,
      "learning_rate": 2.2278641759469477e-06,
      "loss": 0.1697,
      "num_input_tokens_seen": 55626520,
      "step": 34585
    },
    {
      "epoch": 1.565879716607438,
      "grad_norm": 0.36660394072532654,
      "learning_rate": 2.2238146583631825e-06,
      "loss": 0.1167,
      "num_input_tokens_seen": 55634488,
      "step": 34590
    },
    {
      "epoch": 1.5661060684940808,
      "grad_norm": 0.5811144113540649,
      "learning_rate": 2.2197686531534256e-06,
      "loss": 0.1583,
      "num_input_tokens_seen": 55642552,
      "step": 34595
    },
    {
      "epoch": 1.5663324203807238,
      "grad_norm": 0.9100645184516907,
      "learning_rate": 2.2157261609416087e-06,
      "loss": 0.1068,
      "num_input_tokens_seen": 55651352,
      "step": 34600
    },
    {
      "epoch": 1.5663324203807238,
      "eval_loss": 0.14352868497371674,
      "eval_runtime": 404.9205,
      "eval_samples_per_second": 96.984,
      "eval_steps_per_second": 24.247,
      "num_input_tokens_seen": 55651352,
      "step": 34600
    },
    {
      "epoch": 1.5665587722673668,
      "grad_norm": 0.7450835108757019,
      "learning_rate": 2.211687182351149e-06,
      "loss": 0.1388,
      "num_input_tokens_seen": 55659640,
      "step": 34605
    },
    {
      "epoch": 1.5667851241540098,
      "grad_norm": 0.7165966629981995,
      "learning_rate": 2.2076517180048993e-06,
      "loss": 0.0999,
      "num_input_tokens_seen": 55667992,
      "step": 34610
    },
    {
      "epoch": 1.5670114760406528,
      "grad_norm": 0.8078656792640686,
      "learning_rate": 2.2036197685251834e-06,
      "loss": 0.1053,
      "num_input_tokens_seen": 55676664,
      "step": 34615
    },
    {
      "epoch": 1.5672378279272958,
      "grad_norm": 0.31922370195388794,
      "learning_rate": 2.199591334533771e-06,
      "loss": 0.1366,
      "num_input_tokens_seen": 55684696,
      "step": 34620
    },
    {
      "epoch": 1.5674641798139386,
      "grad_norm": 0.5262850522994995,
      "learning_rate": 2.1955664166519036e-06,
      "loss": 0.1607,
      "num_input_tokens_seen": 55692696,
      "step": 34625
    },
    {
      "epoch": 1.5676905317005816,
      "grad_norm": 1.2082130908966064,
      "learning_rate": 2.1915450155002793e-06,
      "loss": 0.1621,
      "num_input_tokens_seen": 55701400,
      "step": 34630
    },
    {
      "epoch": 1.5679168835872246,
      "grad_norm": 0.6403877139091492,
      "learning_rate": 2.187527131699038e-06,
      "loss": 0.1422,
      "num_input_tokens_seen": 55708952,
      "step": 34635
    },
    {
      "epoch": 1.5681432354738676,
      "grad_norm": 0.6210103631019592,
      "learning_rate": 2.18351276586779e-06,
      "loss": 0.1683,
      "num_input_tokens_seen": 55716888,
      "step": 34640
    },
    {
      "epoch": 1.5683695873605106,
      "grad_norm": 0.505411684513092,
      "learning_rate": 2.1795019186256092e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 55725112,
      "step": 34645
    },
    {
      "epoch": 1.5685959392471536,
      "grad_norm": 0.33966103196144104,
      "learning_rate": 2.1754945905910094e-06,
      "loss": 0.165,
      "num_input_tokens_seen": 55732728,
      "step": 34650
    },
    {
      "epoch": 1.5688222911337966,
      "grad_norm": 0.8394870162010193,
      "learning_rate": 2.171490782381977e-06,
      "loss": 0.1558,
      "num_input_tokens_seen": 55740120,
      "step": 34655
    },
    {
      "epoch": 1.5690486430204396,
      "grad_norm": 0.559148907661438,
      "learning_rate": 2.1674904946159425e-06,
      "loss": 0.1348,
      "num_input_tokens_seen": 55747736,
      "step": 34660
    },
    {
      "epoch": 1.5692749949070826,
      "grad_norm": 1.1341098546981812,
      "learning_rate": 2.16349372790981e-06,
      "loss": 0.1562,
      "num_input_tokens_seen": 55755928,
      "step": 34665
    },
    {
      "epoch": 1.5695013467937255,
      "grad_norm": 0.7072052359580994,
      "learning_rate": 2.159500482879928e-06,
      "loss": 0.1363,
      "num_input_tokens_seen": 55763832,
      "step": 34670
    },
    {
      "epoch": 1.5697276986803685,
      "grad_norm": 0.9966477751731873,
      "learning_rate": 2.155510760142096e-06,
      "loss": 0.1771,
      "num_input_tokens_seen": 55771736,
      "step": 34675
    },
    {
      "epoch": 1.5699540505670115,
      "grad_norm": 0.5407899618148804,
      "learning_rate": 2.151524560311588e-06,
      "loss": 0.1211,
      "num_input_tokens_seen": 55779544,
      "step": 34680
    },
    {
      "epoch": 1.5701804024536545,
      "grad_norm": 0.7760618925094604,
      "learning_rate": 2.147541884003129e-06,
      "loss": 0.1678,
      "num_input_tokens_seen": 55787544,
      "step": 34685
    },
    {
      "epoch": 1.5704067543402975,
      "grad_norm": 0.8603341579437256,
      "learning_rate": 2.1435627318308895e-06,
      "loss": 0.1413,
      "num_input_tokens_seen": 55795128,
      "step": 34690
    },
    {
      "epoch": 1.5706331062269405,
      "grad_norm": 0.6493623852729797,
      "learning_rate": 2.139587104408511e-06,
      "loss": 0.1447,
      "num_input_tokens_seen": 55803352,
      "step": 34695
    },
    {
      "epoch": 1.5708594581135835,
      "grad_norm": 0.6689083576202393,
      "learning_rate": 2.1356150023490783e-06,
      "loss": 0.1487,
      "num_input_tokens_seen": 55811128,
      "step": 34700
    },
    {
      "epoch": 1.5710858100002265,
      "grad_norm": 1.0651777982711792,
      "learning_rate": 2.1316464262651464e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 55818776,
      "step": 34705
    },
    {
      "epoch": 1.5713121618868693,
      "grad_norm": 0.33766913414001465,
      "learning_rate": 2.1276813767687224e-06,
      "loss": 0.1418,
      "num_input_tokens_seen": 55827096,
      "step": 34710
    },
    {
      "epoch": 1.5715385137735123,
      "grad_norm": 0.8034429550170898,
      "learning_rate": 2.123719854471254e-06,
      "loss": 0.1112,
      "num_input_tokens_seen": 55835800,
      "step": 34715
    },
    {
      "epoch": 1.5717648656601553,
      "grad_norm": 0.5832423567771912,
      "learning_rate": 2.119761859983668e-06,
      "loss": 0.1085,
      "num_input_tokens_seen": 55843832,
      "step": 34720
    },
    {
      "epoch": 1.5719912175467983,
      "grad_norm": 1.0430899858474731,
      "learning_rate": 2.1158073939163386e-06,
      "loss": 0.1397,
      "num_input_tokens_seen": 55851960,
      "step": 34725
    },
    {
      "epoch": 1.5722175694334413,
      "grad_norm": 0.3163292706012726,
      "learning_rate": 2.111856456879088e-06,
      "loss": 0.1331,
      "num_input_tokens_seen": 55860088,
      "step": 34730
    },
    {
      "epoch": 1.572443921320084,
      "grad_norm": 0.6199480295181274,
      "learning_rate": 2.1079090494811993e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 55868600,
      "step": 34735
    },
    {
      "epoch": 1.572670273206727,
      "grad_norm": 0.5959663391113281,
      "learning_rate": 2.103965172331418e-06,
      "loss": 0.1531,
      "num_input_tokens_seen": 55876632,
      "step": 34740
    },
    {
      "epoch": 1.57289662509337,
      "grad_norm": 1.114682912826538,
      "learning_rate": 2.100024826037933e-06,
      "loss": 0.1425,
      "num_input_tokens_seen": 55885144,
      "step": 34745
    },
    {
      "epoch": 1.573122976980013,
      "grad_norm": 0.5245077013969421,
      "learning_rate": 2.0960880112084027e-06,
      "loss": 0.1855,
      "num_input_tokens_seen": 55893368,
      "step": 34750
    },
    {
      "epoch": 1.573349328866656,
      "grad_norm": 0.7866680026054382,
      "learning_rate": 2.092154728449927e-06,
      "loss": 0.1375,
      "num_input_tokens_seen": 55901624,
      "step": 34755
    },
    {
      "epoch": 1.573575680753299,
      "grad_norm": 0.4451044797897339,
      "learning_rate": 2.0882249783690687e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 55909624,
      "step": 34760
    },
    {
      "epoch": 1.573802032639942,
      "grad_norm": 0.5865907073020935,
      "learning_rate": 2.084298761571851e-06,
      "loss": 0.1246,
      "num_input_tokens_seen": 55917112,
      "step": 34765
    },
    {
      "epoch": 1.574028384526585,
      "grad_norm": 0.9070887565612793,
      "learning_rate": 2.080376078663737e-06,
      "loss": 0.1573,
      "num_input_tokens_seen": 55924952,
      "step": 34770
    },
    {
      "epoch": 1.574254736413228,
      "grad_norm": 0.8933247923851013,
      "learning_rate": 2.0764569302496593e-06,
      "loss": 0.1376,
      "num_input_tokens_seen": 55933048,
      "step": 34775
    },
    {
      "epoch": 1.574481088299871,
      "grad_norm": 0.329750120639801,
      "learning_rate": 2.0725413169339957e-06,
      "loss": 0.0977,
      "num_input_tokens_seen": 55940824,
      "step": 34780
    },
    {
      "epoch": 1.574707440186514,
      "grad_norm": 0.6090821027755737,
      "learning_rate": 2.068629239320588e-06,
      "loss": 0.1521,
      "num_input_tokens_seen": 55949624,
      "step": 34785
    },
    {
      "epoch": 1.574933792073157,
      "grad_norm": 0.8292015790939331,
      "learning_rate": 2.064720698012726e-06,
      "loss": 0.1423,
      "num_input_tokens_seen": 55958168,
      "step": 34790
    },
    {
      "epoch": 1.5751601439598,
      "grad_norm": 0.8006719946861267,
      "learning_rate": 2.0608156936131522e-06,
      "loss": 0.1551,
      "num_input_tokens_seen": 55965592,
      "step": 34795
    },
    {
      "epoch": 1.575386495846443,
      "grad_norm": 0.32651808857917786,
      "learning_rate": 2.056914226724074e-06,
      "loss": 0.1269,
      "num_input_tokens_seen": 55973688,
      "step": 34800
    },
    {
      "epoch": 1.575386495846443,
      "eval_loss": 0.14330258965492249,
      "eval_runtime": 404.7809,
      "eval_samples_per_second": 97.018,
      "eval_steps_per_second": 24.255,
      "num_input_tokens_seen": 55973688,
      "step": 34800
    },
    {
      "epoch": 1.575612847733086,
      "grad_norm": 0.9972383379936218,
      "learning_rate": 2.0530162979471385e-06,
      "loss": 0.1175,
      "num_input_tokens_seen": 55982136,
      "step": 34805
    },
    {
      "epoch": 1.575839199619729,
      "grad_norm": 0.7275519371032715,
      "learning_rate": 2.0491219078834667e-06,
      "loss": 0.1756,
      "num_input_tokens_seen": 55989944,
      "step": 34810
    },
    {
      "epoch": 1.576065551506372,
      "grad_norm": 0.4646754562854767,
      "learning_rate": 2.045231057133612e-06,
      "loss": 0.1105,
      "num_input_tokens_seen": 55998008,
      "step": 34815
    },
    {
      "epoch": 1.5762919033930147,
      "grad_norm": 0.5341973304748535,
      "learning_rate": 2.0413437462975944e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 56005624,
      "step": 34820
    },
    {
      "epoch": 1.5765182552796577,
      "grad_norm": 0.9426380395889282,
      "learning_rate": 2.0374599759748843e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 56013784,
      "step": 34825
    },
    {
      "epoch": 1.5767446071663007,
      "grad_norm": 0.6000601053237915,
      "learning_rate": 2.033579746764419e-06,
      "loss": 0.1691,
      "num_input_tokens_seen": 56022200,
      "step": 34830
    },
    {
      "epoch": 1.5769709590529437,
      "grad_norm": 0.6628919243812561,
      "learning_rate": 2.029703059264565e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 56030040,
      "step": 34835
    },
    {
      "epoch": 1.5771973109395867,
      "grad_norm": 0.47162893414497375,
      "learning_rate": 2.02582991407316e-06,
      "loss": 0.1574,
      "num_input_tokens_seen": 56037784,
      "step": 34840
    },
    {
      "epoch": 1.5774236628262297,
      "grad_norm": 0.7713141441345215,
      "learning_rate": 2.0219603117874992e-06,
      "loss": 0.1452,
      "num_input_tokens_seen": 56046008,
      "step": 34845
    },
    {
      "epoch": 1.5776500147128725,
      "grad_norm": 1.2980064153671265,
      "learning_rate": 2.0180942530043156e-06,
      "loss": 0.1647,
      "num_input_tokens_seen": 56053784,
      "step": 34850
    },
    {
      "epoch": 1.5778763665995155,
      "grad_norm": 0.8064029812812805,
      "learning_rate": 2.0142317383198107e-06,
      "loss": 0.1707,
      "num_input_tokens_seen": 56061944,
      "step": 34855
    },
    {
      "epoch": 1.5781027184861585,
      "grad_norm": 0.8077101707458496,
      "learning_rate": 2.0103727683296243e-06,
      "loss": 0.1592,
      "num_input_tokens_seen": 56070008,
      "step": 34860
    },
    {
      "epoch": 1.5783290703728015,
      "grad_norm": 1.306892991065979,
      "learning_rate": 2.0065173436288636e-06,
      "loss": 0.123,
      "num_input_tokens_seen": 56077784,
      "step": 34865
    },
    {
      "epoch": 1.5785554222594445,
      "grad_norm": 0.8201190233230591,
      "learning_rate": 2.002665464812087e-06,
      "loss": 0.1584,
      "num_input_tokens_seen": 56085304,
      "step": 34870
    },
    {
      "epoch": 1.5787817741460874,
      "grad_norm": 0.6072280406951904,
      "learning_rate": 1.998817132473291e-06,
      "loss": 0.1161,
      "num_input_tokens_seen": 56093784,
      "step": 34875
    },
    {
      "epoch": 1.5790081260327304,
      "grad_norm": 0.884052574634552,
      "learning_rate": 1.9949723472059507e-06,
      "loss": 0.1804,
      "num_input_tokens_seen": 56101400,
      "step": 34880
    },
    {
      "epoch": 1.5792344779193734,
      "grad_norm": 1.2149958610534668,
      "learning_rate": 1.9911311096029726e-06,
      "loss": 0.1294,
      "num_input_tokens_seen": 56109880,
      "step": 34885
    },
    {
      "epoch": 1.5794608298060164,
      "grad_norm": 0.8950063586235046,
      "learning_rate": 1.9872934202567224e-06,
      "loss": 0.1467,
      "num_input_tokens_seen": 56117912,
      "step": 34890
    },
    {
      "epoch": 1.5796871816926594,
      "grad_norm": 0.3684336543083191,
      "learning_rate": 1.9834592797590257e-06,
      "loss": 0.1496,
      "num_input_tokens_seen": 56125784,
      "step": 34895
    },
    {
      "epoch": 1.5799135335793024,
      "grad_norm": 0.4734082520008087,
      "learning_rate": 1.979628688701149e-06,
      "loss": 0.1261,
      "num_input_tokens_seen": 56134136,
      "step": 34900
    },
    {
      "epoch": 1.5801398854659454,
      "grad_norm": 0.762886106967926,
      "learning_rate": 1.9758016476738193e-06,
      "loss": 0.1454,
      "num_input_tokens_seen": 56142200,
      "step": 34905
    },
    {
      "epoch": 1.5803662373525884,
      "grad_norm": 0.5172194242477417,
      "learning_rate": 1.971978157267221e-06,
      "loss": 0.1463,
      "num_input_tokens_seen": 56150072,
      "step": 34910
    },
    {
      "epoch": 1.5805925892392314,
      "grad_norm": 0.9163215160369873,
      "learning_rate": 1.968158218070973e-06,
      "loss": 0.1677,
      "num_input_tokens_seen": 56158520,
      "step": 34915
    },
    {
      "epoch": 1.5808189411258744,
      "grad_norm": 0.6155481338500977,
      "learning_rate": 1.9643418306741682e-06,
      "loss": 0.1674,
      "num_input_tokens_seen": 56166104,
      "step": 34920
    },
    {
      "epoch": 1.5810452930125174,
      "grad_norm": 0.2900611460208893,
      "learning_rate": 1.9605289956653337e-06,
      "loss": 0.1567,
      "num_input_tokens_seen": 56174488,
      "step": 34925
    },
    {
      "epoch": 1.5812716448991604,
      "grad_norm": 0.59190833568573,
      "learning_rate": 1.9567197136324626e-06,
      "loss": 0.123,
      "num_input_tokens_seen": 56183160,
      "step": 34930
    },
    {
      "epoch": 1.5814979967858032,
      "grad_norm": 0.5309609770774841,
      "learning_rate": 1.9529139851629935e-06,
      "loss": 0.131,
      "num_input_tokens_seen": 56191096,
      "step": 34935
    },
    {
      "epoch": 1.5817243486724462,
      "grad_norm": 0.5521154999732971,
      "learning_rate": 1.949111810843812e-06,
      "loss": 0.1413,
      "num_input_tokens_seen": 56199160,
      "step": 34940
    },
    {
      "epoch": 1.5819507005590892,
      "grad_norm": 0.34508901834487915,
      "learning_rate": 1.9453131912612694e-06,
      "loss": 0.1528,
      "num_input_tokens_seen": 56207672,
      "step": 34945
    },
    {
      "epoch": 1.5821770524457321,
      "grad_norm": 0.7545933127403259,
      "learning_rate": 1.941518127001149e-06,
      "loss": 0.159,
      "num_input_tokens_seen": 56216248,
      "step": 34950
    },
    {
      "epoch": 1.5824034043323751,
      "grad_norm": 0.759864091873169,
      "learning_rate": 1.9377266186487107e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 56224024,
      "step": 34955
    },
    {
      "epoch": 1.5826297562190181,
      "grad_norm": 0.44166892766952515,
      "learning_rate": 1.9339386667886483e-06,
      "loss": 0.1382,
      "num_input_tokens_seen": 56232280,
      "step": 34960
    },
    {
      "epoch": 1.582856108105661,
      "grad_norm": 0.517593502998352,
      "learning_rate": 1.9301542720051024e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 56239800,
      "step": 34965
    },
    {
      "epoch": 1.583082459992304,
      "grad_norm": 0.47085142135620117,
      "learning_rate": 1.926373434881684e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 56247576,
      "step": 34970
    },
    {
      "epoch": 1.583308811878947,
      "grad_norm": 0.6781262755393982,
      "learning_rate": 1.9225961560014468e-06,
      "loss": 0.1041,
      "num_input_tokens_seen": 56255960,
      "step": 34975
    },
    {
      "epoch": 1.58353516376559,
      "grad_norm": 0.33610668778419495,
      "learning_rate": 1.918822435946885e-06,
      "loss": 0.1104,
      "num_input_tokens_seen": 56263512,
      "step": 34980
    },
    {
      "epoch": 1.5837615156522329,
      "grad_norm": 0.39649155735969543,
      "learning_rate": 1.915052275299961e-06,
      "loss": 0.1324,
      "num_input_tokens_seen": 56271288,
      "step": 34985
    },
    {
      "epoch": 1.5839878675388759,
      "grad_norm": 0.5653255581855774,
      "learning_rate": 1.9112856746420854e-06,
      "loss": 0.1162,
      "num_input_tokens_seen": 56279256,
      "step": 34990
    },
    {
      "epoch": 1.5842142194255189,
      "grad_norm": 0.33856409788131714,
      "learning_rate": 1.907522634554104e-06,
      "loss": 0.1659,
      "num_input_tokens_seen": 56287448,
      "step": 34995
    },
    {
      "epoch": 1.5844405713121619,
      "grad_norm": 0.3020039498806,
      "learning_rate": 1.9037631556163337e-06,
      "loss": 0.099,
      "num_input_tokens_seen": 56295192,
      "step": 35000
    },
    {
      "epoch": 1.5844405713121619,
      "eval_loss": 0.1436605602502823,
      "eval_runtime": 404.7965,
      "eval_samples_per_second": 97.014,
      "eval_steps_per_second": 24.254,
      "num_input_tokens_seen": 56295192,
      "step": 35000
    },
    {
      "epoch": 1.5846669231988049,
      "grad_norm": 0.7886650562286377,
      "learning_rate": 1.9000072384085272e-06,
      "loss": 0.1231,
      "num_input_tokens_seen": 56303160,
      "step": 35005
    },
    {
      "epoch": 1.5848932750854479,
      "grad_norm": 0.5038684606552124,
      "learning_rate": 1.8962548835098987e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 56311032,
      "step": 35010
    },
    {
      "epoch": 1.5851196269720909,
      "grad_norm": 0.3791996240615845,
      "learning_rate": 1.8925060914991077e-06,
      "loss": 0.1499,
      "num_input_tokens_seen": 56319128,
      "step": 35015
    },
    {
      "epoch": 1.5853459788587339,
      "grad_norm": 0.46467822790145874,
      "learning_rate": 1.888760862954264e-06,
      "loss": 0.1168,
      "num_input_tokens_seen": 56327960,
      "step": 35020
    },
    {
      "epoch": 1.5855723307453768,
      "grad_norm": 0.652256429195404,
      "learning_rate": 1.8850191984529309e-06,
      "loss": 0.1338,
      "num_input_tokens_seen": 56336024,
      "step": 35025
    },
    {
      "epoch": 1.5857986826320198,
      "grad_norm": 0.6544228196144104,
      "learning_rate": 1.8812810985721186e-06,
      "loss": 0.1619,
      "num_input_tokens_seen": 56344568,
      "step": 35030
    },
    {
      "epoch": 1.5860250345186628,
      "grad_norm": 0.6829155683517456,
      "learning_rate": 1.8775465638882856e-06,
      "loss": 0.1435,
      "num_input_tokens_seen": 56352600,
      "step": 35035
    },
    {
      "epoch": 1.5862513864053058,
      "grad_norm": 0.5420457720756531,
      "learning_rate": 1.8738155949773517e-06,
      "loss": 0.1304,
      "num_input_tokens_seen": 56360600,
      "step": 35040
    },
    {
      "epoch": 1.5864777382919488,
      "grad_norm": 1.2424031496047974,
      "learning_rate": 1.8700881924146707e-06,
      "loss": 0.2092,
      "num_input_tokens_seen": 56368824,
      "step": 35045
    },
    {
      "epoch": 1.5867040901785916,
      "grad_norm": 0.8366879820823669,
      "learning_rate": 1.8663643567750577e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 56377016,
      "step": 35050
    },
    {
      "epoch": 1.5869304420652346,
      "grad_norm": 0.91109299659729,
      "learning_rate": 1.8626440886327813e-06,
      "loss": 0.1632,
      "num_input_tokens_seen": 56384856,
      "step": 35055
    },
    {
      "epoch": 1.5871567939518776,
      "grad_norm": 0.48502084612846375,
      "learning_rate": 1.8589273885615432e-06,
      "loss": 0.1368,
      "num_input_tokens_seen": 56392856,
      "step": 35060
    },
    {
      "epoch": 1.5873831458385206,
      "grad_norm": 0.6947983503341675,
      "learning_rate": 1.8552142571345133e-06,
      "loss": 0.1133,
      "num_input_tokens_seen": 56400760,
      "step": 35065
    },
    {
      "epoch": 1.5876094977251636,
      "grad_norm": 0.42355579137802124,
      "learning_rate": 1.8515046949243025e-06,
      "loss": 0.0999,
      "num_input_tokens_seen": 56408824,
      "step": 35070
    },
    {
      "epoch": 1.5878358496118063,
      "grad_norm": 0.8010493516921997,
      "learning_rate": 1.8477987025029674e-06,
      "loss": 0.1316,
      "num_input_tokens_seen": 56417528,
      "step": 35075
    },
    {
      "epoch": 1.5880622014984493,
      "grad_norm": 0.6455453634262085,
      "learning_rate": 1.8440962804420232e-06,
      "loss": 0.0973,
      "num_input_tokens_seen": 56425592,
      "step": 35080
    },
    {
      "epoch": 1.5882885533850923,
      "grad_norm": 0.5120784044265747,
      "learning_rate": 1.8403974293124265e-06,
      "loss": 0.1189,
      "num_input_tokens_seen": 56434168,
      "step": 35085
    },
    {
      "epoch": 1.5885149052717353,
      "grad_norm": 0.8900337219238281,
      "learning_rate": 1.8367021496845854e-06,
      "loss": 0.1517,
      "num_input_tokens_seen": 56442264,
      "step": 35090
    },
    {
      "epoch": 1.5887412571583783,
      "grad_norm": 0.5383245944976807,
      "learning_rate": 1.8330104421283662e-06,
      "loss": 0.1526,
      "num_input_tokens_seen": 56449720,
      "step": 35095
    },
    {
      "epoch": 1.5889676090450213,
      "grad_norm": 1.093033790588379,
      "learning_rate": 1.8293223072130717e-06,
      "loss": 0.1451,
      "num_input_tokens_seen": 56457560,
      "step": 35100
    },
    {
      "epoch": 1.5891939609316643,
      "grad_norm": 0.29470810294151306,
      "learning_rate": 1.8256377455074525e-06,
      "loss": 0.113,
      "num_input_tokens_seen": 56465624,
      "step": 35105
    },
    {
      "epoch": 1.5894203128183073,
      "grad_norm": 0.41389718651771545,
      "learning_rate": 1.8219567575797263e-06,
      "loss": 0.1213,
      "num_input_tokens_seen": 56474104,
      "step": 35110
    },
    {
      "epoch": 1.5896466647049503,
      "grad_norm": 0.20933912694454193,
      "learning_rate": 1.8182793439975365e-06,
      "loss": 0.1131,
      "num_input_tokens_seen": 56482392,
      "step": 35115
    },
    {
      "epoch": 1.5898730165915933,
      "grad_norm": 0.8692532181739807,
      "learning_rate": 1.8146055053279958e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 56489912,
      "step": 35120
    },
    {
      "epoch": 1.5900993684782363,
      "grad_norm": 0.7348572611808777,
      "learning_rate": 1.8109352421376486e-06,
      "loss": 0.1371,
      "num_input_tokens_seen": 56497656,
      "step": 35125
    },
    {
      "epoch": 1.5903257203648793,
      "grad_norm": 1.1444042921066284,
      "learning_rate": 1.8072685549924972e-06,
      "loss": 0.1617,
      "num_input_tokens_seen": 56505816,
      "step": 35130
    },
    {
      "epoch": 1.5905520722515223,
      "grad_norm": 0.7958723306655884,
      "learning_rate": 1.8036054444579982e-06,
      "loss": 0.1329,
      "num_input_tokens_seen": 56513752,
      "step": 35135
    },
    {
      "epoch": 1.5907784241381653,
      "grad_norm": 0.8225705623626709,
      "learning_rate": 1.7999459110990407e-06,
      "loss": 0.1142,
      "num_input_tokens_seen": 56521560,
      "step": 35140
    },
    {
      "epoch": 1.5910047760248083,
      "grad_norm": 0.9746513366699219,
      "learning_rate": 1.7962899554799712e-06,
      "loss": 0.1372,
      "num_input_tokens_seen": 56529592,
      "step": 35145
    },
    {
      "epoch": 1.5912311279114513,
      "grad_norm": 0.7262024283409119,
      "learning_rate": 1.7926375781645937e-06,
      "loss": 0.1557,
      "num_input_tokens_seen": 56537176,
      "step": 35150
    },
    {
      "epoch": 1.5914574797980943,
      "grad_norm": 0.46695396304130554,
      "learning_rate": 1.7889887797161359e-06,
      "loss": 0.1321,
      "num_input_tokens_seen": 56545752,
      "step": 35155
    },
    {
      "epoch": 1.591683831684737,
      "grad_norm": 0.6148775219917297,
      "learning_rate": 1.7853435606973028e-06,
      "loss": 0.1382,
      "num_input_tokens_seen": 56553688,
      "step": 35160
    },
    {
      "epoch": 1.59191018357138,
      "grad_norm": 0.807414710521698,
      "learning_rate": 1.781701921670223e-06,
      "loss": 0.1699,
      "num_input_tokens_seen": 56561880,
      "step": 35165
    },
    {
      "epoch": 1.592136535458023,
      "grad_norm": 0.9650975465774536,
      "learning_rate": 1.7780638631964886e-06,
      "loss": 0.1412,
      "num_input_tokens_seen": 56569944,
      "step": 35170
    },
    {
      "epoch": 1.592362887344666,
      "grad_norm": 1.1916872262954712,
      "learning_rate": 1.7744293858371314e-06,
      "loss": 0.1534,
      "num_input_tokens_seen": 56578168,
      "step": 35175
    },
    {
      "epoch": 1.592589239231309,
      "grad_norm": 0.5710080862045288,
      "learning_rate": 1.770798490152631e-06,
      "loss": 0.1381,
      "num_input_tokens_seen": 56586040,
      "step": 35180
    },
    {
      "epoch": 1.592815591117952,
      "grad_norm": 0.8571557998657227,
      "learning_rate": 1.767171176702917e-06,
      "loss": 0.1342,
      "num_input_tokens_seen": 56594072,
      "step": 35185
    },
    {
      "epoch": 1.5930419430045948,
      "grad_norm": 0.3763003349304199,
      "learning_rate": 1.7635474460473755e-06,
      "loss": 0.1624,
      "num_input_tokens_seen": 56602008,
      "step": 35190
    },
    {
      "epoch": 1.5932682948912378,
      "grad_norm": 0.72749263048172,
      "learning_rate": 1.7599272987448206e-06,
      "loss": 0.1315,
      "num_input_tokens_seen": 56610200,
      "step": 35195
    },
    {
      "epoch": 1.5934946467778808,
      "grad_norm": 0.7279271483421326,
      "learning_rate": 1.7563107353535362e-06,
      "loss": 0.1574,
      "num_input_tokens_seen": 56617656,
      "step": 35200
    },
    {
      "epoch": 1.5934946467778808,
      "eval_loss": 0.1435719132423401,
      "eval_runtime": 405.2727,
      "eval_samples_per_second": 96.9,
      "eval_steps_per_second": 24.226,
      "num_input_tokens_seen": 56617656,
      "step": 35200
    },
    {
      "epoch": 1.5937209986645238,
      "grad_norm": 0.7604751586914062,
      "learning_rate": 1.7526977564312263e-06,
      "loss": 0.1539,
      "num_input_tokens_seen": 56625368,
      "step": 35205
    },
    {
      "epoch": 1.5939473505511668,
      "grad_norm": 0.3771846294403076,
      "learning_rate": 1.7490883625350701e-06,
      "loss": 0.1373,
      "num_input_tokens_seen": 56633048,
      "step": 35210
    },
    {
      "epoch": 1.5941737024378098,
      "grad_norm": 1.4462039470672607,
      "learning_rate": 1.7454825542216807e-06,
      "loss": 0.1723,
      "num_input_tokens_seen": 56641304,
      "step": 35215
    },
    {
      "epoch": 1.5944000543244528,
      "grad_norm": 0.5253553986549377,
      "learning_rate": 1.7418803320471105e-06,
      "loss": 0.1526,
      "num_input_tokens_seen": 56649016,
      "step": 35220
    },
    {
      "epoch": 1.5946264062110957,
      "grad_norm": 0.5411932468414307,
      "learning_rate": 1.7382816965668737e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 56657144,
      "step": 35225
    },
    {
      "epoch": 1.5948527580977387,
      "grad_norm": 0.36161965131759644,
      "learning_rate": 1.7346866483359285e-06,
      "loss": 0.1573,
      "num_input_tokens_seen": 56665432,
      "step": 35230
    },
    {
      "epoch": 1.5950791099843817,
      "grad_norm": 0.8711602687835693,
      "learning_rate": 1.7310951879086657e-06,
      "loss": 0.1508,
      "num_input_tokens_seen": 56673016,
      "step": 35235
    },
    {
      "epoch": 1.5953054618710247,
      "grad_norm": 0.35712066292762756,
      "learning_rate": 1.7275073158389471e-06,
      "loss": 0.1659,
      "num_input_tokens_seen": 56680984,
      "step": 35240
    },
    {
      "epoch": 1.5955318137576677,
      "grad_norm": 1.1137274503707886,
      "learning_rate": 1.723923032680061e-06,
      "loss": 0.1766,
      "num_input_tokens_seen": 56689400,
      "step": 35245
    },
    {
      "epoch": 1.5957581656443107,
      "grad_norm": 0.5320279002189636,
      "learning_rate": 1.7203423389847428e-06,
      "loss": 0.1825,
      "num_input_tokens_seen": 56697688,
      "step": 35250
    },
    {
      "epoch": 1.5959845175309537,
      "grad_norm": 0.7609041333198547,
      "learning_rate": 1.7167652353051928e-06,
      "loss": 0.1606,
      "num_input_tokens_seen": 56705784,
      "step": 35255
    },
    {
      "epoch": 1.5962108694175967,
      "grad_norm": 0.37036192417144775,
      "learning_rate": 1.7131917221930333e-06,
      "loss": 0.1502,
      "num_input_tokens_seen": 56713080,
      "step": 35260
    },
    {
      "epoch": 1.5964372213042397,
      "grad_norm": 0.5977098345756531,
      "learning_rate": 1.7096218001993513e-06,
      "loss": 0.1585,
      "num_input_tokens_seen": 56721240,
      "step": 35265
    },
    {
      "epoch": 1.5966635731908827,
      "grad_norm": 0.34271255135536194,
      "learning_rate": 1.706055469874676e-06,
      "loss": 0.1387,
      "num_input_tokens_seen": 56729592,
      "step": 35270
    },
    {
      "epoch": 1.5968899250775255,
      "grad_norm": 0.6471575498580933,
      "learning_rate": 1.702492731768976e-06,
      "loss": 0.1698,
      "num_input_tokens_seen": 56739320,
      "step": 35275
    },
    {
      "epoch": 1.5971162769641685,
      "grad_norm": 0.457412987947464,
      "learning_rate": 1.6989335864316724e-06,
      "loss": 0.1224,
      "num_input_tokens_seen": 56747192,
      "step": 35280
    },
    {
      "epoch": 1.5973426288508115,
      "grad_norm": 0.47048646211624146,
      "learning_rate": 1.6953780344116265e-06,
      "loss": 0.1643,
      "num_input_tokens_seen": 56755192,
      "step": 35285
    },
    {
      "epoch": 1.5975689807374545,
      "grad_norm": 0.4959375262260437,
      "learning_rate": 1.6918260762571497e-06,
      "loss": 0.1098,
      "num_input_tokens_seen": 56763256,
      "step": 35290
    },
    {
      "epoch": 1.5977953326240975,
      "grad_norm": 0.3333672285079956,
      "learning_rate": 1.6882777125160093e-06,
      "loss": 0.1529,
      "num_input_tokens_seen": 56771704,
      "step": 35295
    },
    {
      "epoch": 1.5980216845107404,
      "grad_norm": 0.43256717920303345,
      "learning_rate": 1.6847329437353899e-06,
      "loss": 0.1244,
      "num_input_tokens_seen": 56779640,
      "step": 35300
    },
    {
      "epoch": 1.5982480363973832,
      "grad_norm": 0.8450986742973328,
      "learning_rate": 1.6811917704619511e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 56787448,
      "step": 35305
    },
    {
      "epoch": 1.5984743882840262,
      "grad_norm": 0.3332269787788391,
      "learning_rate": 1.67765419324179e-06,
      "loss": 0.1066,
      "num_input_tokens_seen": 56795256,
      "step": 35310
    },
    {
      "epoch": 1.5987007401706692,
      "grad_norm": 0.6239641308784485,
      "learning_rate": 1.6741202126204364e-06,
      "loss": 0.1527,
      "num_input_tokens_seen": 56803032,
      "step": 35315
    },
    {
      "epoch": 1.5989270920573122,
      "grad_norm": 0.8090705871582031,
      "learning_rate": 1.6705898291428767e-06,
      "loss": 0.1501,
      "num_input_tokens_seen": 56810680,
      "step": 35320
    },
    {
      "epoch": 1.5991534439439552,
      "grad_norm": 0.7301941514015198,
      "learning_rate": 1.6670630433535395e-06,
      "loss": 0.1629,
      "num_input_tokens_seen": 56819256,
      "step": 35325
    },
    {
      "epoch": 1.5993797958305982,
      "grad_norm": 0.5296999216079712,
      "learning_rate": 1.6635398557962979e-06,
      "loss": 0.1017,
      "num_input_tokens_seen": 56827000,
      "step": 35330
    },
    {
      "epoch": 1.5996061477172412,
      "grad_norm": 0.3101922273635864,
      "learning_rate": 1.660020267014481e-06,
      "loss": 0.1351,
      "num_input_tokens_seen": 56834648,
      "step": 35335
    },
    {
      "epoch": 1.5998324996038842,
      "grad_norm": 0.9614182114601135,
      "learning_rate": 1.6565042775508438e-06,
      "loss": 0.18,
      "num_input_tokens_seen": 56842808,
      "step": 35340
    },
    {
      "epoch": 1.6000588514905272,
      "grad_norm": 1.3720191717147827,
      "learning_rate": 1.6529918879475997e-06,
      "loss": 0.1436,
      "num_input_tokens_seen": 56850840,
      "step": 35345
    },
    {
      "epoch": 1.6002852033771702,
      "grad_norm": 0.38394665718078613,
      "learning_rate": 1.6494830987464043e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 56858744,
      "step": 35350
    },
    {
      "epoch": 1.6005115552638132,
      "grad_norm": 0.5942526459693909,
      "learning_rate": 1.6459779104883555e-06,
      "loss": 0.1256,
      "num_input_tokens_seen": 56866680,
      "step": 35355
    },
    {
      "epoch": 1.6007379071504562,
      "grad_norm": 0.5847994089126587,
      "learning_rate": 1.6424763237140013e-06,
      "loss": 0.1292,
      "num_input_tokens_seen": 56874040,
      "step": 35360
    },
    {
      "epoch": 1.6009642590370992,
      "grad_norm": 0.27051135897636414,
      "learning_rate": 1.6389783389633207e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 56881912,
      "step": 35365
    },
    {
      "epoch": 1.6011906109237422,
      "grad_norm": 0.6052566170692444,
      "learning_rate": 1.6354839567757546e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 56890040,
      "step": 35370
    },
    {
      "epoch": 1.6014169628103851,
      "grad_norm": 1.3092528581619263,
      "learning_rate": 1.6319931776901831e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 56898264,
      "step": 35375
    },
    {
      "epoch": 1.6016433146970281,
      "grad_norm": 0.7081972360610962,
      "learning_rate": 1.6285060022449229e-06,
      "loss": 0.1295,
      "num_input_tokens_seen": 56906360,
      "step": 35380
    },
    {
      "epoch": 1.6018696665836711,
      "grad_norm": 0.7557283639907837,
      "learning_rate": 1.6250224309777434e-06,
      "loss": 0.1298,
      "num_input_tokens_seen": 56914776,
      "step": 35385
    },
    {
      "epoch": 1.602096018470314,
      "grad_norm": 0.29880931973457336,
      "learning_rate": 1.6215424644258515e-06,
      "loss": 0.1123,
      "num_input_tokens_seen": 56922328,
      "step": 35390
    },
    {
      "epoch": 1.602322370356957,
      "grad_norm": 0.8381667137145996,
      "learning_rate": 1.6180661031259036e-06,
      "loss": 0.1118,
      "num_input_tokens_seen": 56930648,
      "step": 35395
    },
    {
      "epoch": 1.6025487222436,
      "grad_norm": 0.7138304710388184,
      "learning_rate": 1.614593347613999e-06,
      "loss": 0.1345,
      "num_input_tokens_seen": 56938168,
      "step": 35400
    },
    {
      "epoch": 1.6025487222436,
      "eval_loss": 0.14333750307559967,
      "eval_runtime": 404.684,
      "eval_samples_per_second": 97.041,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 56938168,
      "step": 35400
    },
    {
      "epoch": 1.602775074130243,
      "grad_norm": 1.2218436002731323,
      "learning_rate": 1.6111241984256758e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 56947448,
      "step": 35405
    },
    {
      "epoch": 1.6030014260168859,
      "grad_norm": 0.4315662682056427,
      "learning_rate": 1.6076586560959257e-06,
      "loss": 0.1175,
      "num_input_tokens_seen": 56955032,
      "step": 35410
    },
    {
      "epoch": 1.6032277779035287,
      "grad_norm": 0.4886932969093323,
      "learning_rate": 1.604196721159182e-06,
      "loss": 0.1058,
      "num_input_tokens_seen": 56963096,
      "step": 35415
    },
    {
      "epoch": 1.6034541297901717,
      "grad_norm": 0.8764874935150146,
      "learning_rate": 1.6007383941493092e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 56970904,
      "step": 35420
    },
    {
      "epoch": 1.6036804816768147,
      "grad_norm": 1.2784861326217651,
      "learning_rate": 1.5972836755996285e-06,
      "loss": 0.1492,
      "num_input_tokens_seen": 56979384,
      "step": 35425
    },
    {
      "epoch": 1.6039068335634576,
      "grad_norm": 1.0333516597747803,
      "learning_rate": 1.5938325660429076e-06,
      "loss": 0.1043,
      "num_input_tokens_seen": 56987704,
      "step": 35430
    },
    {
      "epoch": 1.6041331854501006,
      "grad_norm": 0.8090140223503113,
      "learning_rate": 1.5903850660113378e-06,
      "loss": 0.1239,
      "num_input_tokens_seen": 56995384,
      "step": 35435
    },
    {
      "epoch": 1.6043595373367436,
      "grad_norm": 0.5310725569725037,
      "learning_rate": 1.5869411760365826e-06,
      "loss": 0.1638,
      "num_input_tokens_seen": 57003608,
      "step": 35440
    },
    {
      "epoch": 1.6045858892233866,
      "grad_norm": 0.513472318649292,
      "learning_rate": 1.58350089664972e-06,
      "loss": 0.1299,
      "num_input_tokens_seen": 57011608,
      "step": 35445
    },
    {
      "epoch": 1.6048122411100296,
      "grad_norm": 1.0769433975219727,
      "learning_rate": 1.5800642283812865e-06,
      "loss": 0.1336,
      "num_input_tokens_seen": 57019864,
      "step": 35450
    },
    {
      "epoch": 1.6050385929966726,
      "grad_norm": 0.4752112627029419,
      "learning_rate": 1.5766311717612698e-06,
      "loss": 0.1204,
      "num_input_tokens_seen": 57027736,
      "step": 35455
    },
    {
      "epoch": 1.6052649448833156,
      "grad_norm": 0.40752366185188293,
      "learning_rate": 1.5732017273190818e-06,
      "loss": 0.1328,
      "num_input_tokens_seen": 57036056,
      "step": 35460
    },
    {
      "epoch": 1.6054912967699586,
      "grad_norm": 0.4375614821910858,
      "learning_rate": 1.5697758955835806e-06,
      "loss": 0.1882,
      "num_input_tokens_seen": 57043832,
      "step": 35465
    },
    {
      "epoch": 1.6057176486566016,
      "grad_norm": 0.2779771089553833,
      "learning_rate": 1.566353677083085e-06,
      "loss": 0.1668,
      "num_input_tokens_seen": 57051448,
      "step": 35470
    },
    {
      "epoch": 1.6059440005432446,
      "grad_norm": 1.1212880611419678,
      "learning_rate": 1.562935072345334e-06,
      "loss": 0.1428,
      "num_input_tokens_seen": 57059544,
      "step": 35475
    },
    {
      "epoch": 1.6061703524298876,
      "grad_norm": 1.3172342777252197,
      "learning_rate": 1.5595200818975281e-06,
      "loss": 0.1401,
      "num_input_tokens_seen": 57067448,
      "step": 35480
    },
    {
      "epoch": 1.6063967043165306,
      "grad_norm": 0.8955281972885132,
      "learning_rate": 1.5561087062662905e-06,
      "loss": 0.1296,
      "num_input_tokens_seen": 57075736,
      "step": 35485
    },
    {
      "epoch": 1.6066230562031736,
      "grad_norm": 0.3007262945175171,
      "learning_rate": 1.5527009459777087e-06,
      "loss": 0.1238,
      "num_input_tokens_seen": 57084376,
      "step": 35490
    },
    {
      "epoch": 1.6068494080898166,
      "grad_norm": 0.5502066612243652,
      "learning_rate": 1.5492968015572984e-06,
      "loss": 0.1292,
      "num_input_tokens_seen": 57092088,
      "step": 35495
    },
    {
      "epoch": 1.6070757599764594,
      "grad_norm": 0.9664018154144287,
      "learning_rate": 1.5458962735300203e-06,
      "loss": 0.1531,
      "num_input_tokens_seen": 57100440,
      "step": 35500
    },
    {
      "epoch": 1.6073021118631023,
      "grad_norm": 0.2856472432613373,
      "learning_rate": 1.54249936242028e-06,
      "loss": 0.0881,
      "num_input_tokens_seen": 57108056,
      "step": 35505
    },
    {
      "epoch": 1.6075284637497453,
      "grad_norm": 0.42122703790664673,
      "learning_rate": 1.5391060687519222e-06,
      "loss": 0.1253,
      "num_input_tokens_seen": 57115832,
      "step": 35510
    },
    {
      "epoch": 1.6077548156363883,
      "grad_norm": 0.6311815977096558,
      "learning_rate": 1.5357163930482367e-06,
      "loss": 0.1282,
      "num_input_tokens_seen": 57124312,
      "step": 35515
    },
    {
      "epoch": 1.6079811675230313,
      "grad_norm": 0.9184883832931519,
      "learning_rate": 1.532330335831955e-06,
      "loss": 0.1471,
      "num_input_tokens_seen": 57133080,
      "step": 35520
    },
    {
      "epoch": 1.6082075194096743,
      "grad_norm": 0.465600848197937,
      "learning_rate": 1.5289478976252491e-06,
      "loss": 0.1601,
      "num_input_tokens_seen": 57141336,
      "step": 35525
    },
    {
      "epoch": 1.608433871296317,
      "grad_norm": 0.6580111980438232,
      "learning_rate": 1.5255690789497345e-06,
      "loss": 0.1188,
      "num_input_tokens_seen": 57149112,
      "step": 35530
    },
    {
      "epoch": 1.60866022318296,
      "grad_norm": 0.42960914969444275,
      "learning_rate": 1.5221938803264641e-06,
      "loss": 0.1151,
      "num_input_tokens_seen": 57156664,
      "step": 35535
    },
    {
      "epoch": 1.608886575069603,
      "grad_norm": 1.1276273727416992,
      "learning_rate": 1.518822302275938e-06,
      "loss": 0.18,
      "num_input_tokens_seen": 57164760,
      "step": 35540
    },
    {
      "epoch": 1.609112926956246,
      "grad_norm": 0.5350795388221741,
      "learning_rate": 1.5154543453180958e-06,
      "loss": 0.1299,
      "num_input_tokens_seen": 57172920,
      "step": 35545
    },
    {
      "epoch": 1.609339278842889,
      "grad_norm": 0.4664972424507141,
      "learning_rate": 1.5120900099723167e-06,
      "loss": 0.1266,
      "num_input_tokens_seen": 57180760,
      "step": 35550
    },
    {
      "epoch": 1.609565630729532,
      "grad_norm": 0.29595819115638733,
      "learning_rate": 1.5087292967574273e-06,
      "loss": 0.1168,
      "num_input_tokens_seen": 57188632,
      "step": 35555
    },
    {
      "epoch": 1.609791982616175,
      "grad_norm": 1.033705472946167,
      "learning_rate": 1.5053722061916908e-06,
      "loss": 0.2018,
      "num_input_tokens_seen": 57196664,
      "step": 35560
    },
    {
      "epoch": 1.610018334502818,
      "grad_norm": 0.41145059466362,
      "learning_rate": 1.5020187387928124e-06,
      "loss": 0.139,
      "num_input_tokens_seen": 57204696,
      "step": 35565
    },
    {
      "epoch": 1.610244686389461,
      "grad_norm": 0.4555746614933014,
      "learning_rate": 1.4986688950779343e-06,
      "loss": 0.1315,
      "num_input_tokens_seen": 57212664,
      "step": 35570
    },
    {
      "epoch": 1.610471038276104,
      "grad_norm": 0.645340621471405,
      "learning_rate": 1.495322675563654e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 57220728,
      "step": 35575
    },
    {
      "epoch": 1.610697390162747,
      "grad_norm": 0.6262407302856445,
      "learning_rate": 1.4919800807659922e-06,
      "loss": 0.1213,
      "num_input_tokens_seen": 57228376,
      "step": 35580
    },
    {
      "epoch": 1.61092374204939,
      "grad_norm": 0.4638485312461853,
      "learning_rate": 1.4886411112004255e-06,
      "loss": 0.1279,
      "num_input_tokens_seen": 57237112,
      "step": 35585
    },
    {
      "epoch": 1.611150093936033,
      "grad_norm": 0.7616966962814331,
      "learning_rate": 1.4853057673818588e-06,
      "loss": 0.173,
      "num_input_tokens_seen": 57245272,
      "step": 35590
    },
    {
      "epoch": 1.611376445822676,
      "grad_norm": 1.019834280014038,
      "learning_rate": 1.481974049824647e-06,
      "loss": 0.1195,
      "num_input_tokens_seen": 57253112,
      "step": 35595
    },
    {
      "epoch": 1.611602797709319,
      "grad_norm": 1.3073862791061401,
      "learning_rate": 1.4786459590425849e-06,
      "loss": 0.1602,
      "num_input_tokens_seen": 57260536,
      "step": 35600
    },
    {
      "epoch": 1.611602797709319,
      "eval_loss": 0.14349424839019775,
      "eval_runtime": 405.2359,
      "eval_samples_per_second": 96.909,
      "eval_steps_per_second": 24.228,
      "num_input_tokens_seen": 57260536,
      "step": 35600
    },
    {
      "epoch": 1.611829149595962,
      "grad_norm": 0.6067479848861694,
      "learning_rate": 1.4753214955489036e-06,
      "loss": 0.1622,
      "num_input_tokens_seen": 57268152,
      "step": 35605
    },
    {
      "epoch": 1.612055501482605,
      "grad_norm": 0.9123795032501221,
      "learning_rate": 1.4720006598562737e-06,
      "loss": 0.1304,
      "num_input_tokens_seen": 57276408,
      "step": 35610
    },
    {
      "epoch": 1.6122818533692478,
      "grad_norm": 0.954525887966156,
      "learning_rate": 1.4686834524768185e-06,
      "loss": 0.1485,
      "num_input_tokens_seen": 57285304,
      "step": 35615
    },
    {
      "epoch": 1.6125082052558908,
      "grad_norm": 0.24352003633975983,
      "learning_rate": 1.4653698739220844e-06,
      "loss": 0.1079,
      "num_input_tokens_seen": 57293368,
      "step": 35620
    },
    {
      "epoch": 1.6127345571425338,
      "grad_norm": 0.44026392698287964,
      "learning_rate": 1.4620599247030715e-06,
      "loss": 0.1085,
      "num_input_tokens_seen": 57301784,
      "step": 35625
    },
    {
      "epoch": 1.6129609090291768,
      "grad_norm": 0.5154592394828796,
      "learning_rate": 1.4587536053302125e-06,
      "loss": 0.1531,
      "num_input_tokens_seen": 57309432,
      "step": 35630
    },
    {
      "epoch": 1.6131872609158198,
      "grad_norm": 0.42647668719291687,
      "learning_rate": 1.4554509163133862e-06,
      "loss": 0.1477,
      "num_input_tokens_seen": 57317304,
      "step": 35635
    },
    {
      "epoch": 1.6134136128024628,
      "grad_norm": 0.8109027743339539,
      "learning_rate": 1.4521518581619098e-06,
      "loss": 0.1631,
      "num_input_tokens_seen": 57325528,
      "step": 35640
    },
    {
      "epoch": 1.6136399646891055,
      "grad_norm": 0.5934911966323853,
      "learning_rate": 1.4488564313845348e-06,
      "loss": 0.129,
      "num_input_tokens_seen": 57333944,
      "step": 35645
    },
    {
      "epoch": 1.6138663165757485,
      "grad_norm": 1.0301358699798584,
      "learning_rate": 1.4455646364894603e-06,
      "loss": 0.1964,
      "num_input_tokens_seen": 57341432,
      "step": 35650
    },
    {
      "epoch": 1.6140926684623915,
      "grad_norm": 0.9978863596916199,
      "learning_rate": 1.4422764739843247e-06,
      "loss": 0.1377,
      "num_input_tokens_seen": 57350040,
      "step": 35655
    },
    {
      "epoch": 1.6143190203490345,
      "grad_norm": 0.31369468569755554,
      "learning_rate": 1.4389919443762e-06,
      "loss": 0.1378,
      "num_input_tokens_seen": 57357816,
      "step": 35660
    },
    {
      "epoch": 1.6145453722356775,
      "grad_norm": 0.23484081029891968,
      "learning_rate": 1.4357110481716063e-06,
      "loss": 0.1175,
      "num_input_tokens_seen": 57365816,
      "step": 35665
    },
    {
      "epoch": 1.6147717241223205,
      "grad_norm": 0.32630711793899536,
      "learning_rate": 1.4324337858764941e-06,
      "loss": 0.1399,
      "num_input_tokens_seen": 57373880,
      "step": 35670
    },
    {
      "epoch": 1.6149980760089635,
      "grad_norm": 0.5504868626594543,
      "learning_rate": 1.4291601579962622e-06,
      "loss": 0.1551,
      "num_input_tokens_seen": 57381528,
      "step": 35675
    },
    {
      "epoch": 1.6152244278956065,
      "grad_norm": 0.9153584241867065,
      "learning_rate": 1.42589016503574e-06,
      "loss": 0.107,
      "num_input_tokens_seen": 57389496,
      "step": 35680
    },
    {
      "epoch": 1.6154507797822495,
      "grad_norm": 0.8343351483345032,
      "learning_rate": 1.4226238074992099e-06,
      "loss": 0.1498,
      "num_input_tokens_seen": 57396984,
      "step": 35685
    },
    {
      "epoch": 1.6156771316688925,
      "grad_norm": 1.1216442584991455,
      "learning_rate": 1.4193610858903778e-06,
      "loss": 0.1513,
      "num_input_tokens_seen": 57405336,
      "step": 35690
    },
    {
      "epoch": 1.6159034835555355,
      "grad_norm": 1.094544529914856,
      "learning_rate": 1.416102000712402e-06,
      "loss": 0.1503,
      "num_input_tokens_seen": 57414264,
      "step": 35695
    },
    {
      "epoch": 1.6161298354421785,
      "grad_norm": 0.33695900440216064,
      "learning_rate": 1.4128465524678668e-06,
      "loss": 0.111,
      "num_input_tokens_seen": 57421976,
      "step": 35700
    },
    {
      "epoch": 1.6163561873288215,
      "grad_norm": 0.5011320114135742,
      "learning_rate": 1.4095947416588124e-06,
      "loss": 0.1458,
      "num_input_tokens_seen": 57430104,
      "step": 35705
    },
    {
      "epoch": 1.6165825392154645,
      "grad_norm": 0.5277765989303589,
      "learning_rate": 1.4063465687866983e-06,
      "loss": 0.1194,
      "num_input_tokens_seen": 57438264,
      "step": 35710
    },
    {
      "epoch": 1.6168088911021075,
      "grad_norm": 0.24256955087184906,
      "learning_rate": 1.4031020343524438e-06,
      "loss": 0.1228,
      "num_input_tokens_seen": 57446296,
      "step": 35715
    },
    {
      "epoch": 1.6170352429887505,
      "grad_norm": 0.47754284739494324,
      "learning_rate": 1.3998611388563926e-06,
      "loss": 0.144,
      "num_input_tokens_seen": 57454296,
      "step": 35720
    },
    {
      "epoch": 1.6172615948753934,
      "grad_norm": 0.4421539604663849,
      "learning_rate": 1.3966238827983314e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 57462264,
      "step": 35725
    },
    {
      "epoch": 1.6174879467620362,
      "grad_norm": 0.5906452536582947,
      "learning_rate": 1.393390266677483e-06,
      "loss": 0.1632,
      "num_input_tokens_seen": 57470296,
      "step": 35730
    },
    {
      "epoch": 1.6177142986486792,
      "grad_norm": 0.431963711977005,
      "learning_rate": 1.3901602909925204e-06,
      "loss": 0.1244,
      "num_input_tokens_seen": 57477656,
      "step": 35735
    },
    {
      "epoch": 1.6179406505353222,
      "grad_norm": 0.2026100903749466,
      "learning_rate": 1.3869339562415373e-06,
      "loss": 0.0914,
      "num_input_tokens_seen": 57486040,
      "step": 35740
    },
    {
      "epoch": 1.6181670024219652,
      "grad_norm": 0.7769182920455933,
      "learning_rate": 1.38371126292208e-06,
      "loss": 0.1487,
      "num_input_tokens_seen": 57493560,
      "step": 35745
    },
    {
      "epoch": 1.6183933543086082,
      "grad_norm": 0.6340448260307312,
      "learning_rate": 1.3804922115311286e-06,
      "loss": 0.1495,
      "num_input_tokens_seen": 57501720,
      "step": 35750
    },
    {
      "epoch": 1.618619706195251,
      "grad_norm": 0.5123820304870605,
      "learning_rate": 1.3772768025650945e-06,
      "loss": 0.1062,
      "num_input_tokens_seen": 57510424,
      "step": 35755
    },
    {
      "epoch": 1.618846058081894,
      "grad_norm": 1.1760985851287842,
      "learning_rate": 1.3740650365198448e-06,
      "loss": 0.1738,
      "num_input_tokens_seen": 57518584,
      "step": 35760
    },
    {
      "epoch": 1.619072409968537,
      "grad_norm": 1.3425979614257812,
      "learning_rate": 1.3708569138906612e-06,
      "loss": 0.1356,
      "num_input_tokens_seen": 57526520,
      "step": 35765
    },
    {
      "epoch": 1.61929876185518,
      "grad_norm": 0.7785547971725464,
      "learning_rate": 1.367652435172287e-06,
      "loss": 0.1542,
      "num_input_tokens_seen": 57534264,
      "step": 35770
    },
    {
      "epoch": 1.619525113741823,
      "grad_norm": 0.6974373459815979,
      "learning_rate": 1.364451600858893e-06,
      "loss": 0.1295,
      "num_input_tokens_seen": 57543032,
      "step": 35775
    },
    {
      "epoch": 1.619751465628466,
      "grad_norm": 1.2722111940383911,
      "learning_rate": 1.3612544114440823e-06,
      "loss": 0.1347,
      "num_input_tokens_seen": 57551224,
      "step": 35780
    },
    {
      "epoch": 1.619977817515109,
      "grad_norm": 0.7303276658058167,
      "learning_rate": 1.3580608674209072e-06,
      "loss": 0.1455,
      "num_input_tokens_seen": 57558520,
      "step": 35785
    },
    {
      "epoch": 1.620204169401752,
      "grad_norm": 0.5892728567123413,
      "learning_rate": 1.3548709692818434e-06,
      "loss": 0.1703,
      "num_input_tokens_seen": 57566168,
      "step": 35790
    },
    {
      "epoch": 1.620430521288395,
      "grad_norm": 0.7229293584823608,
      "learning_rate": 1.3516847175188223e-06,
      "loss": 0.1474,
      "num_input_tokens_seen": 57574072,
      "step": 35795
    },
    {
      "epoch": 1.620656873175038,
      "grad_norm": 0.3824661672115326,
      "learning_rate": 1.348502112623204e-06,
      "loss": 0.136,
      "num_input_tokens_seen": 57581848,
      "step": 35800
    },
    {
      "epoch": 1.620656873175038,
      "eval_loss": 0.1434459090232849,
      "eval_runtime": 404.3748,
      "eval_samples_per_second": 97.115,
      "eval_steps_per_second": 24.279,
      "num_input_tokens_seen": 57581848,
      "step": 35800
    },
    {
      "epoch": 1.620883225061681,
      "grad_norm": 0.3410244286060333,
      "learning_rate": 1.3453231550857787e-06,
      "loss": 0.1139,
      "num_input_tokens_seen": 57589848,
      "step": 35805
    },
    {
      "epoch": 1.621109576948324,
      "grad_norm": 0.9795024991035461,
      "learning_rate": 1.3421478453967878e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 57598232,
      "step": 35810
    },
    {
      "epoch": 1.621335928834967,
      "grad_norm": 0.4129793345928192,
      "learning_rate": 1.3389761840459065e-06,
      "loss": 0.1086,
      "num_input_tokens_seen": 57606232,
      "step": 35815
    },
    {
      "epoch": 1.62156228072161,
      "grad_norm": 1.2789044380187988,
      "learning_rate": 1.3358081715222376e-06,
      "loss": 0.1386,
      "num_input_tokens_seen": 57614712,
      "step": 35820
    },
    {
      "epoch": 1.621788632608253,
      "grad_norm": 0.9076271057128906,
      "learning_rate": 1.3326438083143295e-06,
      "loss": 0.1726,
      "num_input_tokens_seen": 57622648,
      "step": 35825
    },
    {
      "epoch": 1.622014984494896,
      "grad_norm": 0.8049716353416443,
      "learning_rate": 1.3294830949101723e-06,
      "loss": 0.1327,
      "num_input_tokens_seen": 57630808,
      "step": 35830
    },
    {
      "epoch": 1.622241336381539,
      "grad_norm": 1.5987502336502075,
      "learning_rate": 1.3263260317971815e-06,
      "loss": 0.12,
      "num_input_tokens_seen": 57639288,
      "step": 35835
    },
    {
      "epoch": 1.6224676882681817,
      "grad_norm": 0.43435853719711304,
      "learning_rate": 1.3231726194622208e-06,
      "loss": 0.1263,
      "num_input_tokens_seen": 57646584,
      "step": 35840
    },
    {
      "epoch": 1.6226940401548247,
      "grad_norm": 0.39459601044654846,
      "learning_rate": 1.3200228583915814e-06,
      "loss": 0.1617,
      "num_input_tokens_seen": 57654712,
      "step": 35845
    },
    {
      "epoch": 1.6229203920414677,
      "grad_norm": 0.819712221622467,
      "learning_rate": 1.3168767490709971e-06,
      "loss": 0.096,
      "num_input_tokens_seen": 57662360,
      "step": 35850
    },
    {
      "epoch": 1.6231467439281106,
      "grad_norm": 0.5378835797309875,
      "learning_rate": 1.3137342919856437e-06,
      "loss": 0.1593,
      "num_input_tokens_seen": 57670392,
      "step": 35855
    },
    {
      "epoch": 1.6233730958147536,
      "grad_norm": 0.48277151584625244,
      "learning_rate": 1.310595487620117e-06,
      "loss": 0.1384,
      "num_input_tokens_seen": 57678008,
      "step": 35860
    },
    {
      "epoch": 1.6235994477013966,
      "grad_norm": 0.4026508927345276,
      "learning_rate": 1.3074603364584715e-06,
      "loss": 0.1284,
      "num_input_tokens_seen": 57685944,
      "step": 35865
    },
    {
      "epoch": 1.6238257995880394,
      "grad_norm": 0.724582314491272,
      "learning_rate": 1.3043288389841758e-06,
      "loss": 0.1693,
      "num_input_tokens_seen": 57694200,
      "step": 35870
    },
    {
      "epoch": 1.6240521514746824,
      "grad_norm": 0.6769068837165833,
      "learning_rate": 1.3012009956801546e-06,
      "loss": 0.1395,
      "num_input_tokens_seen": 57701816,
      "step": 35875
    },
    {
      "epoch": 1.6242785033613254,
      "grad_norm": 0.29839617013931274,
      "learning_rate": 1.2980768070287586e-06,
      "loss": 0.1355,
      "num_input_tokens_seen": 57709880,
      "step": 35880
    },
    {
      "epoch": 1.6245048552479684,
      "grad_norm": 0.4055117666721344,
      "learning_rate": 1.2949562735117716e-06,
      "loss": 0.1157,
      "num_input_tokens_seen": 57717752,
      "step": 35885
    },
    {
      "epoch": 1.6247312071346114,
      "grad_norm": 0.4592527747154236,
      "learning_rate": 1.291839395610428e-06,
      "loss": 0.1339,
      "num_input_tokens_seen": 57726616,
      "step": 35890
    },
    {
      "epoch": 1.6249575590212544,
      "grad_norm": 0.6961389780044556,
      "learning_rate": 1.2887261738053852e-06,
      "loss": 0.1398,
      "num_input_tokens_seen": 57734264,
      "step": 35895
    },
    {
      "epoch": 1.6251839109078974,
      "grad_norm": 0.7277741432189941,
      "learning_rate": 1.2856166085767396e-06,
      "loss": 0.1504,
      "num_input_tokens_seen": 57742168,
      "step": 35900
    },
    {
      "epoch": 1.6254102627945404,
      "grad_norm": 0.5103989839553833,
      "learning_rate": 1.2825107004040272e-06,
      "loss": 0.1068,
      "num_input_tokens_seen": 57749752,
      "step": 35905
    },
    {
      "epoch": 1.6256366146811834,
      "grad_norm": 0.5282511115074158,
      "learning_rate": 1.2794084497662146e-06,
      "loss": 0.1221,
      "num_input_tokens_seen": 57757688,
      "step": 35910
    },
    {
      "epoch": 1.6258629665678264,
      "grad_norm": 0.7298911213874817,
      "learning_rate": 1.276309857141711e-06,
      "loss": 0.1386,
      "num_input_tokens_seen": 57765784,
      "step": 35915
    },
    {
      "epoch": 1.6260893184544694,
      "grad_norm": 1.2085224390029907,
      "learning_rate": 1.273214923008359e-06,
      "loss": 0.1316,
      "num_input_tokens_seen": 57773656,
      "step": 35920
    },
    {
      "epoch": 1.6263156703411124,
      "grad_norm": 0.41539472341537476,
      "learning_rate": 1.2701236478434352e-06,
      "loss": 0.1915,
      "num_input_tokens_seen": 57781080,
      "step": 35925
    },
    {
      "epoch": 1.6265420222277553,
      "grad_norm": 0.42108023166656494,
      "learning_rate": 1.2670360321236502e-06,
      "loss": 0.1219,
      "num_input_tokens_seen": 57789112,
      "step": 35930
    },
    {
      "epoch": 1.6267683741143983,
      "grad_norm": 0.6720402836799622,
      "learning_rate": 1.2639520763251617e-06,
      "loss": 0.1132,
      "num_input_tokens_seen": 57796984,
      "step": 35935
    },
    {
      "epoch": 1.6269947260010413,
      "grad_norm": 0.47793060541152954,
      "learning_rate": 1.2608717809235448e-06,
      "loss": 0.1464,
      "num_input_tokens_seen": 57805176,
      "step": 35940
    },
    {
      "epoch": 1.6272210778876843,
      "grad_norm": 0.5815293788909912,
      "learning_rate": 1.2577951463938282e-06,
      "loss": 0.1674,
      "num_input_tokens_seen": 57813656,
      "step": 35945
    },
    {
      "epoch": 1.6274474297743273,
      "grad_norm": 0.7082526087760925,
      "learning_rate": 1.2547221732104569e-06,
      "loss": 0.1674,
      "num_input_tokens_seen": 57821400,
      "step": 35950
    },
    {
      "epoch": 1.62767378166097,
      "grad_norm": 1.0153911113739014,
      "learning_rate": 1.25165286184733e-06,
      "loss": 0.1332,
      "num_input_tokens_seen": 57829112,
      "step": 35955
    },
    {
      "epoch": 1.627900133547613,
      "grad_norm": 0.3029613494873047,
      "learning_rate": 1.248587212777777e-06,
      "loss": 0.1558,
      "num_input_tokens_seen": 57837048,
      "step": 35960
    },
    {
      "epoch": 1.628126485434256,
      "grad_norm": 0.2979200482368469,
      "learning_rate": 1.2455252264745532e-06,
      "loss": 0.1331,
      "num_input_tokens_seen": 57844600,
      "step": 35965
    },
    {
      "epoch": 1.628352837320899,
      "grad_norm": 0.3336913287639618,
      "learning_rate": 1.2424669034098528e-06,
      "loss": 0.1597,
      "num_input_tokens_seen": 57852664,
      "step": 35970
    },
    {
      "epoch": 1.628579189207542,
      "grad_norm": 0.7707982659339905,
      "learning_rate": 1.2394122440553185e-06,
      "loss": 0.1334,
      "num_input_tokens_seen": 57860472,
      "step": 35975
    },
    {
      "epoch": 1.628805541094185,
      "grad_norm": 0.2907412648200989,
      "learning_rate": 1.2363612488820037e-06,
      "loss": 0.1335,
      "num_input_tokens_seen": 57868472,
      "step": 35980
    },
    {
      "epoch": 1.6290318929808278,
      "grad_norm": 0.4421588182449341,
      "learning_rate": 1.2333139183604208e-06,
      "loss": 0.1394,
      "num_input_tokens_seen": 57876248,
      "step": 35985
    },
    {
      "epoch": 1.6292582448674708,
      "grad_norm": 1.2707617282867432,
      "learning_rate": 1.2302702529604998e-06,
      "loss": 0.1822,
      "num_input_tokens_seen": 57883960,
      "step": 35990
    },
    {
      "epoch": 1.6294845967541138,
      "grad_norm": 0.5274625420570374,
      "learning_rate": 1.227230253151615e-06,
      "loss": 0.1576,
      "num_input_tokens_seen": 57892952,
      "step": 35995
    },
    {
      "epoch": 1.6297109486407568,
      "grad_norm": 0.7856091260910034,
      "learning_rate": 1.2241939194025748e-06,
      "loss": 0.1444,
      "num_input_tokens_seen": 57900760,
      "step": 36000
    },
    {
      "epoch": 1.6297109486407568,
      "eval_loss": 0.14369554817676544,
      "eval_runtime": 405.0341,
      "eval_samples_per_second": 96.957,
      "eval_steps_per_second": 24.24,
      "num_input_tokens_seen": 57900760,
      "step": 36000
    },
    {
      "epoch": 1.6299373005273998,
      "grad_norm": 0.3621406853199005,
      "learning_rate": 1.2211612521816156e-06,
      "loss": 0.1315,
      "num_input_tokens_seen": 57908344,
      "step": 36005
    },
    {
      "epoch": 1.6301636524140428,
      "grad_norm": 0.47156640887260437,
      "learning_rate": 1.2181322519564137e-06,
      "loss": 0.1449,
      "num_input_tokens_seen": 57916024,
      "step": 36010
    },
    {
      "epoch": 1.6303900043006858,
      "grad_norm": 0.4624112844467163,
      "learning_rate": 1.2151069191940839e-06,
      "loss": 0.1403,
      "num_input_tokens_seen": 57924344,
      "step": 36015
    },
    {
      "epoch": 1.6306163561873288,
      "grad_norm": 0.7837782502174377,
      "learning_rate": 1.2120852543611644e-06,
      "loss": 0.1484,
      "num_input_tokens_seen": 57932152,
      "step": 36020
    },
    {
      "epoch": 1.6308427080739718,
      "grad_norm": 0.4341008961200714,
      "learning_rate": 1.2090672579236379e-06,
      "loss": 0.1354,
      "num_input_tokens_seen": 57940376,
      "step": 36025
    },
    {
      "epoch": 1.6310690599606148,
      "grad_norm": 0.4319596588611603,
      "learning_rate": 1.2060529303469126e-06,
      "loss": 0.1599,
      "num_input_tokens_seen": 57948408,
      "step": 36030
    },
    {
      "epoch": 1.6312954118472578,
      "grad_norm": 1.3806626796722412,
      "learning_rate": 1.2030422720958445e-06,
      "loss": 0.1754,
      "num_input_tokens_seen": 57956312,
      "step": 36035
    },
    {
      "epoch": 1.6315217637339008,
      "grad_norm": 1.1459826231002808,
      "learning_rate": 1.200035283634704e-06,
      "loss": 0.1636,
      "num_input_tokens_seen": 57964376,
      "step": 36040
    },
    {
      "epoch": 1.6317481156205438,
      "grad_norm": 0.9792083501815796,
      "learning_rate": 1.1970319654272144e-06,
      "loss": 0.108,
      "num_input_tokens_seen": 57972280,
      "step": 36045
    },
    {
      "epoch": 1.6319744675071868,
      "grad_norm": 0.5622294545173645,
      "learning_rate": 1.1940323179365192e-06,
      "loss": 0.1046,
      "num_input_tokens_seen": 57980248,
      "step": 36050
    },
    {
      "epoch": 1.6322008193938298,
      "grad_norm": 0.7781476974487305,
      "learning_rate": 1.1910363416252095e-06,
      "loss": 0.1589,
      "num_input_tokens_seen": 57989080,
      "step": 36055
    },
    {
      "epoch": 1.6324271712804728,
      "grad_norm": 0.4564252495765686,
      "learning_rate": 1.1880440369552964e-06,
      "loss": 0.151,
      "num_input_tokens_seen": 57997336,
      "step": 36060
    },
    {
      "epoch": 1.6326535231671158,
      "grad_norm": 0.3898526728153229,
      "learning_rate": 1.1850554043882328e-06,
      "loss": 0.1,
      "num_input_tokens_seen": 58005464,
      "step": 36065
    },
    {
      "epoch": 1.6328798750537585,
      "grad_norm": 0.8952118754386902,
      "learning_rate": 1.1820704443849028e-06,
      "loss": 0.1577,
      "num_input_tokens_seen": 58014136,
      "step": 36070
    },
    {
      "epoch": 1.6331062269404015,
      "grad_norm": 0.8464339375495911,
      "learning_rate": 1.1790891574056219e-06,
      "loss": 0.2053,
      "num_input_tokens_seen": 58021784,
      "step": 36075
    },
    {
      "epoch": 1.6333325788270445,
      "grad_norm": 1.2142765522003174,
      "learning_rate": 1.1761115439101523e-06,
      "loss": 0.1306,
      "num_input_tokens_seen": 58030712,
      "step": 36080
    },
    {
      "epoch": 1.6335589307136875,
      "grad_norm": 0.6717212796211243,
      "learning_rate": 1.1731376043576659e-06,
      "loss": 0.1903,
      "num_input_tokens_seen": 58038936,
      "step": 36085
    },
    {
      "epoch": 1.6337852826003305,
      "grad_norm": 0.9374292492866516,
      "learning_rate": 1.1701673392067875e-06,
      "loss": 0.1392,
      "num_input_tokens_seen": 58046584,
      "step": 36090
    },
    {
      "epoch": 1.6340116344869733,
      "grad_norm": 0.36365628242492676,
      "learning_rate": 1.1672007489155757e-06,
      "loss": 0.1545,
      "num_input_tokens_seen": 58054968,
      "step": 36095
    },
    {
      "epoch": 1.6342379863736163,
      "grad_norm": 0.5320423245429993,
      "learning_rate": 1.164237833941506e-06,
      "loss": 0.1212,
      "num_input_tokens_seen": 58062776,
      "step": 36100
    },
    {
      "epoch": 1.6344643382602593,
      "grad_norm": 0.4346444606781006,
      "learning_rate": 1.1612785947415022e-06,
      "loss": 0.1404,
      "num_input_tokens_seen": 58070904,
      "step": 36105
    },
    {
      "epoch": 1.6346906901469023,
      "grad_norm": 0.48498252034187317,
      "learning_rate": 1.1583230317719185e-06,
      "loss": 0.1291,
      "num_input_tokens_seen": 58078712,
      "step": 36110
    },
    {
      "epoch": 1.6349170420335453,
      "grad_norm": 0.5394761562347412,
      "learning_rate": 1.1553711454885318e-06,
      "loss": 0.161,
      "num_input_tokens_seen": 58087704,
      "step": 36115
    },
    {
      "epoch": 1.6351433939201883,
      "grad_norm": 1.3385350704193115,
      "learning_rate": 1.152422936346567e-06,
      "loss": 0.1909,
      "num_input_tokens_seen": 58095416,
      "step": 36120
    },
    {
      "epoch": 1.6353697458068313,
      "grad_norm": 0.46487176418304443,
      "learning_rate": 1.1494784048006718e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 58103768,
      "step": 36125
    },
    {
      "epoch": 1.6355960976934742,
      "grad_norm": 0.7330628037452698,
      "learning_rate": 1.1465375513049326e-06,
      "loss": 0.1377,
      "num_input_tokens_seen": 58111608,
      "step": 36130
    },
    {
      "epoch": 1.6358224495801172,
      "grad_norm": 0.3045976758003235,
      "learning_rate": 1.1436003763128616e-06,
      "loss": 0.1564,
      "num_input_tokens_seen": 58119192,
      "step": 36135
    },
    {
      "epoch": 1.6360488014667602,
      "grad_norm": 0.6169305443763733,
      "learning_rate": 1.1406668802774106e-06,
      "loss": 0.1694,
      "num_input_tokens_seen": 58128472,
      "step": 36140
    },
    {
      "epoch": 1.6362751533534032,
      "grad_norm": 0.7987351417541504,
      "learning_rate": 1.137737063650965e-06,
      "loss": 0.1307,
      "num_input_tokens_seen": 58137016,
      "step": 36145
    },
    {
      "epoch": 1.6365015052400462,
      "grad_norm": 0.5172485113143921,
      "learning_rate": 1.1348109268853323e-06,
      "loss": 0.1577,
      "num_input_tokens_seen": 58145144,
      "step": 36150
    },
    {
      "epoch": 1.6367278571266892,
      "grad_norm": 0.3317907452583313,
      "learning_rate": 1.1318884704317634e-06,
      "loss": 0.1598,
      "num_input_tokens_seen": 58153112,
      "step": 36155
    },
    {
      "epoch": 1.6369542090133322,
      "grad_norm": 1.1028553247451782,
      "learning_rate": 1.1289696947409417e-06,
      "loss": 0.1787,
      "num_input_tokens_seen": 58161176,
      "step": 36160
    },
    {
      "epoch": 1.6371805608999752,
      "grad_norm": 0.5156553387641907,
      "learning_rate": 1.126054600262974e-06,
      "loss": 0.1616,
      "num_input_tokens_seen": 58169432,
      "step": 36165
    },
    {
      "epoch": 1.6374069127866182,
      "grad_norm": 1.4554699659347534,
      "learning_rate": 1.1231431874474064e-06,
      "loss": 0.172,
      "num_input_tokens_seen": 58177432,
      "step": 36170
    },
    {
      "epoch": 1.6376332646732612,
      "grad_norm": 0.8500785231590271,
      "learning_rate": 1.12023545674321e-06,
      "loss": 0.1518,
      "num_input_tokens_seen": 58186424,
      "step": 36175
    },
    {
      "epoch": 1.637859616559904,
      "grad_norm": 1.2016398906707764,
      "learning_rate": 1.117331408598804e-06,
      "loss": 0.1289,
      "num_input_tokens_seen": 58194200,
      "step": 36180
    },
    {
      "epoch": 1.638085968446547,
      "grad_norm": 0.6514430642127991,
      "learning_rate": 1.1144310434620191e-06,
      "loss": 0.1225,
      "num_input_tokens_seen": 58201912,
      "step": 36185
    },
    {
      "epoch": 1.63831232033319,
      "grad_norm": 0.76140296459198,
      "learning_rate": 1.1115343617801365e-06,
      "loss": 0.1586,
      "num_input_tokens_seen": 58209944,
      "step": 36190
    },
    {
      "epoch": 1.638538672219833,
      "grad_norm": 0.7118211984634399,
      "learning_rate": 1.1086413639998515e-06,
      "loss": 0.1416,
      "num_input_tokens_seen": 58217880,
      "step": 36195
    },
    {
      "epoch": 1.638765024106476,
      "grad_norm": 0.7948864698410034,
      "learning_rate": 1.1057520505673103e-06,
      "loss": 0.1473,
      "num_input_tokens_seen": 58226104,
      "step": 36200
    },
    {
      "epoch": 1.638765024106476,
      "eval_loss": 0.14405041933059692,
      "eval_runtime": 404.4279,
      "eval_samples_per_second": 97.103,
      "eval_steps_per_second": 24.276,
      "num_input_tokens_seen": 58226104,
      "step": 36200
    },
    {
      "epoch": 1.638991375993119,
      "grad_norm": 0.3779137432575226,
      "learning_rate": 1.1028664219280727e-06,
      "loss": 0.1128,
      "num_input_tokens_seen": 58234488,
      "step": 36205
    },
    {
      "epoch": 1.6392177278797617,
      "grad_norm": 1.3392999172210693,
      "learning_rate": 1.0999844785271468e-06,
      "loss": 0.1469,
      "num_input_tokens_seen": 58242168,
      "step": 36210
    },
    {
      "epoch": 1.6394440797664047,
      "grad_norm": 0.4356326460838318,
      "learning_rate": 1.097106220808955e-06,
      "loss": 0.1436,
      "num_input_tokens_seen": 58249752,
      "step": 36215
    },
    {
      "epoch": 1.6396704316530477,
      "grad_norm": 0.9924771189689636,
      "learning_rate": 1.0942316492173698e-06,
      "loss": 0.1338,
      "num_input_tokens_seen": 58257752,
      "step": 36220
    },
    {
      "epoch": 1.6398967835396907,
      "grad_norm": 0.4924844205379486,
      "learning_rate": 1.0913607641956841e-06,
      "loss": 0.1668,
      "num_input_tokens_seen": 58265368,
      "step": 36225
    },
    {
      "epoch": 1.6401231354263337,
      "grad_norm": 0.5633832812309265,
      "learning_rate": 1.0884935661866213e-06,
      "loss": 0.1287,
      "num_input_tokens_seen": 58273112,
      "step": 36230
    },
    {
      "epoch": 1.6403494873129767,
      "grad_norm": 0.5581543445587158,
      "learning_rate": 1.0856300556323418e-06,
      "loss": 0.119,
      "num_input_tokens_seen": 58280920,
      "step": 36235
    },
    {
      "epoch": 1.6405758391996197,
      "grad_norm": 0.7237072587013245,
      "learning_rate": 1.0827702329744365e-06,
      "loss": 0.1291,
      "num_input_tokens_seen": 58289208,
      "step": 36240
    },
    {
      "epoch": 1.6408021910862627,
      "grad_norm": 0.40798041224479675,
      "learning_rate": 1.0799140986539197e-06,
      "loss": 0.1595,
      "num_input_tokens_seen": 58297240,
      "step": 36245
    },
    {
      "epoch": 1.6410285429729057,
      "grad_norm": 0.4279024004936218,
      "learning_rate": 1.0770616531112526e-06,
      "loss": 0.1376,
      "num_input_tokens_seen": 58304920,
      "step": 36250
    },
    {
      "epoch": 1.6412548948595487,
      "grad_norm": 0.8358420729637146,
      "learning_rate": 1.0742128967863085e-06,
      "loss": 0.183,
      "num_input_tokens_seen": 58312440,
      "step": 36255
    },
    {
      "epoch": 1.6414812467461917,
      "grad_norm": 0.4071051776409149,
      "learning_rate": 1.071367830118411e-06,
      "loss": 0.1255,
      "num_input_tokens_seen": 58320280,
      "step": 36260
    },
    {
      "epoch": 1.6417075986328347,
      "grad_norm": 0.5110769271850586,
      "learning_rate": 1.068526453546298e-06,
      "loss": 0.1252,
      "num_input_tokens_seen": 58328504,
      "step": 36265
    },
    {
      "epoch": 1.6419339505194777,
      "grad_norm": 0.6054601073265076,
      "learning_rate": 1.0656887675081467e-06,
      "loss": 0.1508,
      "num_input_tokens_seen": 58337560,
      "step": 36270
    },
    {
      "epoch": 1.6421603024061207,
      "grad_norm": 0.9556020498275757,
      "learning_rate": 1.0628547724415628e-06,
      "loss": 0.1444,
      "num_input_tokens_seen": 58346104,
      "step": 36275
    },
    {
      "epoch": 1.6423866542927636,
      "grad_norm": 0.5241594910621643,
      "learning_rate": 1.0600244687835881e-06,
      "loss": 0.1587,
      "num_input_tokens_seen": 58354168,
      "step": 36280
    },
    {
      "epoch": 1.6426130061794066,
      "grad_norm": 0.5836312770843506,
      "learning_rate": 1.0571978569706876e-06,
      "loss": 0.1255,
      "num_input_tokens_seen": 58362584,
      "step": 36285
    },
    {
      "epoch": 1.6428393580660496,
      "grad_norm": 0.5429351329803467,
      "learning_rate": 1.0543749374387652e-06,
      "loss": 0.106,
      "num_input_tokens_seen": 58370776,
      "step": 36290
    },
    {
      "epoch": 1.6430657099526924,
      "grad_norm": 0.8502735495567322,
      "learning_rate": 1.051555710623142e-06,
      "loss": 0.156,
      "num_input_tokens_seen": 58378360,
      "step": 36295
    },
    {
      "epoch": 1.6432920618393354,
      "grad_norm": 0.8739970326423645,
      "learning_rate": 1.0487401769585847e-06,
      "loss": 0.1258,
      "num_input_tokens_seen": 58386424,
      "step": 36300
    },
    {
      "epoch": 1.6435184137259784,
      "grad_norm": 0.9945608377456665,
      "learning_rate": 1.0459283368792845e-06,
      "loss": 0.1507,
      "num_input_tokens_seen": 58395192,
      "step": 36305
    },
    {
      "epoch": 1.6437447656126214,
      "grad_norm": 0.8020422458648682,
      "learning_rate": 1.043120190818858e-06,
      "loss": 0.1509,
      "num_input_tokens_seen": 58403704,
      "step": 36310
    },
    {
      "epoch": 1.6439711174992644,
      "grad_norm": 1.2897675037384033,
      "learning_rate": 1.0403157392103596e-06,
      "loss": 0.1837,
      "num_input_tokens_seen": 58411960,
      "step": 36315
    },
    {
      "epoch": 1.6441974693859074,
      "grad_norm": 0.7870146632194519,
      "learning_rate": 1.0375149824862735e-06,
      "loss": 0.1526,
      "num_input_tokens_seen": 58420088,
      "step": 36320
    },
    {
      "epoch": 1.6444238212725502,
      "grad_norm": 0.5732016563415527,
      "learning_rate": 1.034717921078507e-06,
      "loss": 0.1282,
      "num_input_tokens_seen": 58428024,
      "step": 36325
    },
    {
      "epoch": 1.6446501731591932,
      "grad_norm": 0.8643355369567871,
      "learning_rate": 1.0319245554184009e-06,
      "loss": 0.1325,
      "num_input_tokens_seen": 58435832,
      "step": 36330
    },
    {
      "epoch": 1.6448765250458361,
      "grad_norm": 0.3014513850212097,
      "learning_rate": 1.0291348859367361e-06,
      "loss": 0.1045,
      "num_input_tokens_seen": 58443608,
      "step": 36335
    },
    {
      "epoch": 1.6451028769324791,
      "grad_norm": 0.6095685958862305,
      "learning_rate": 1.0263489130637016e-06,
      "loss": 0.1463,
      "num_input_tokens_seen": 58451800,
      "step": 36340
    },
    {
      "epoch": 1.6453292288191221,
      "grad_norm": 0.4235992133617401,
      "learning_rate": 1.0235666372289427e-06,
      "loss": 0.152,
      "num_input_tokens_seen": 58459256,
      "step": 36345
    },
    {
      "epoch": 1.6455555807057651,
      "grad_norm": 0.8464927673339844,
      "learning_rate": 1.0207880588615076e-06,
      "loss": 0.1434,
      "num_input_tokens_seen": 58467640,
      "step": 36350
    },
    {
      "epoch": 1.6457819325924081,
      "grad_norm": 0.422020822763443,
      "learning_rate": 1.0180131783898984e-06,
      "loss": 0.1393,
      "num_input_tokens_seen": 58475832,
      "step": 36355
    },
    {
      "epoch": 1.6460082844790511,
      "grad_norm": 1.4456685781478882,
      "learning_rate": 1.0152419962420362e-06,
      "loss": 0.1736,
      "num_input_tokens_seen": 58483416,
      "step": 36360
    },
    {
      "epoch": 1.6462346363656941,
      "grad_norm": 0.4839628338813782,
      "learning_rate": 1.0124745128452685e-06,
      "loss": 0.143,
      "num_input_tokens_seen": 58491544,
      "step": 36365
    },
    {
      "epoch": 1.646460988252337,
      "grad_norm": 1.1184355020523071,
      "learning_rate": 1.0097107286263758e-06,
      "loss": 0.1465,
      "num_input_tokens_seen": 58499896,
      "step": 36370
    },
    {
      "epoch": 1.64668734013898,
      "grad_norm": 0.5986302495002747,
      "learning_rate": 1.00695064401157e-06,
      "loss": 0.1406,
      "num_input_tokens_seen": 58507800,
      "step": 36375
    },
    {
      "epoch": 1.646913692025623,
      "grad_norm": 0.86024409532547,
      "learning_rate": 1.0041942594264886e-06,
      "loss": 0.1297,
      "num_input_tokens_seen": 58516088,
      "step": 36380
    },
    {
      "epoch": 1.647140043912266,
      "grad_norm": 0.3559160828590393,
      "learning_rate": 1.001441575296208e-06,
      "loss": 0.1374,
      "num_input_tokens_seen": 58523896,
      "step": 36385
    },
    {
      "epoch": 1.647366395798909,
      "grad_norm": 0.700171947479248,
      "learning_rate": 9.986925920452139e-07,
      "loss": 0.1122,
      "num_input_tokens_seen": 58531832,
      "step": 36390
    },
    {
      "epoch": 1.647592747685552,
      "grad_norm": 0.6029492616653442,
      "learning_rate": 9.959473100974475e-07,
      "loss": 0.1447,
      "num_input_tokens_seen": 58539928,
      "step": 36395
    },
    {
      "epoch": 1.647819099572195,
      "grad_norm": 0.6840978860855103,
      "learning_rate": 9.932057298762564e-07,
      "loss": 0.1587,
      "num_input_tokens_seen": 58548024,
      "step": 36400
    },
    {
      "epoch": 1.647819099572195,
      "eval_loss": 0.1433379054069519,
      "eval_runtime": 404.6691,
      "eval_samples_per_second": 97.045,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 58548024,
      "step": 36400
    },
    {
      "epoch": 1.648045451458838,
      "grad_norm": 0.5398637652397156,
      "learning_rate": 9.90467851804433e-07,
      "loss": 0.1389,
      "num_input_tokens_seen": 58555992,
      "step": 36405
    },
    {
      "epoch": 1.6482718033454808,
      "grad_norm": 0.4522177278995514,
      "learning_rate": 9.877336763041895e-07,
      "loss": 0.14,
      "num_input_tokens_seen": 58564024,
      "step": 36410
    },
    {
      "epoch": 1.6484981552321238,
      "grad_norm": 0.6831315755844116,
      "learning_rate": 9.850032037971662e-07,
      "loss": 0.1141,
      "num_input_tokens_seen": 58572792,
      "step": 36415
    },
    {
      "epoch": 1.6487245071187668,
      "grad_norm": 0.4998997747898102,
      "learning_rate": 9.822764347044406e-07,
      "loss": 0.1562,
      "num_input_tokens_seen": 58580728,
      "step": 36420
    },
    {
      "epoch": 1.6489508590054098,
      "grad_norm": 0.3649001121520996,
      "learning_rate": 9.795533694465175e-07,
      "loss": 0.1599,
      "num_input_tokens_seen": 58589080,
      "step": 36425
    },
    {
      "epoch": 1.6491772108920528,
      "grad_norm": 0.7696269154548645,
      "learning_rate": 9.768340084433197e-07,
      "loss": 0.1491,
      "num_input_tokens_seen": 58596856,
      "step": 36430
    },
    {
      "epoch": 1.6494035627786956,
      "grad_norm": 0.37916576862335205,
      "learning_rate": 9.741183521142143e-07,
      "loss": 0.1394,
      "num_input_tokens_seen": 58604280,
      "step": 36435
    },
    {
      "epoch": 1.6496299146653386,
      "grad_norm": 0.7019485831260681,
      "learning_rate": 9.714064008779889e-07,
      "loss": 0.1619,
      "num_input_tokens_seen": 58612344,
      "step": 36440
    },
    {
      "epoch": 1.6498562665519816,
      "grad_norm": 1.1322928667068481,
      "learning_rate": 9.686981551528584e-07,
      "loss": 0.1895,
      "num_input_tokens_seen": 58620760,
      "step": 36445
    },
    {
      "epoch": 1.6500826184386246,
      "grad_norm": 0.8855730295181274,
      "learning_rate": 9.65993615356467e-07,
      "loss": 0.156,
      "num_input_tokens_seen": 58628568,
      "step": 36450
    },
    {
      "epoch": 1.6503089703252676,
      "grad_norm": 0.6015740633010864,
      "learning_rate": 9.632927819058917e-07,
      "loss": 0.1439,
      "num_input_tokens_seen": 58636856,
      "step": 36455
    },
    {
      "epoch": 1.6505353222119106,
      "grad_norm": 0.6515997052192688,
      "learning_rate": 9.605956552176305e-07,
      "loss": 0.1546,
      "num_input_tokens_seen": 58646488,
      "step": 36460
    },
    {
      "epoch": 1.6507616740985536,
      "grad_norm": 0.2720901072025299,
      "learning_rate": 9.579022357076223e-07,
      "loss": 0.1378,
      "num_input_tokens_seen": 58654808,
      "step": 36465
    },
    {
      "epoch": 1.6509880259851966,
      "grad_norm": 0.681557834148407,
      "learning_rate": 9.552125237912158e-07,
      "loss": 0.1363,
      "num_input_tokens_seen": 58663128,
      "step": 36470
    },
    {
      "epoch": 1.6512143778718396,
      "grad_norm": 0.45263150334358215,
      "learning_rate": 9.525265198832096e-07,
      "loss": 0.1466,
      "num_input_tokens_seen": 58670648,
      "step": 36475
    },
    {
      "epoch": 1.6514407297584826,
      "grad_norm": 0.6284640431404114,
      "learning_rate": 9.498442243978112e-07,
      "loss": 0.1141,
      "num_input_tokens_seen": 58677816,
      "step": 36480
    },
    {
      "epoch": 1.6516670816451255,
      "grad_norm": 0.5697517395019531,
      "learning_rate": 9.471656377486649e-07,
      "loss": 0.1501,
      "num_input_tokens_seen": 58685784,
      "step": 36485
    },
    {
      "epoch": 1.6518934335317685,
      "grad_norm": 0.48468971252441406,
      "learning_rate": 9.444907603488456e-07,
      "loss": 0.1593,
      "num_input_tokens_seen": 58694136,
      "step": 36490
    },
    {
      "epoch": 1.6521197854184115,
      "grad_norm": 0.38565483689308167,
      "learning_rate": 9.418195926108514e-07,
      "loss": 0.141,
      "num_input_tokens_seen": 58701720,
      "step": 36495
    },
    {
      "epoch": 1.6523461373050545,
      "grad_norm": 0.2644543945789337,
      "learning_rate": 9.391521349466053e-07,
      "loss": 0.1325,
      "num_input_tokens_seen": 58709912,
      "step": 36500
    },
    {
      "epoch": 1.6525724891916975,
      "grad_norm": 0.4447438418865204,
      "learning_rate": 9.364883877674758e-07,
      "loss": 0.1725,
      "num_input_tokens_seen": 58718072,
      "step": 36505
    },
    {
      "epoch": 1.6527988410783405,
      "grad_norm": 1.1134088039398193,
      "learning_rate": 9.33828351484231e-07,
      "loss": 0.1436,
      "num_input_tokens_seen": 58726136,
      "step": 36510
    },
    {
      "epoch": 1.6530251929649835,
      "grad_norm": 1.0686135292053223,
      "learning_rate": 9.311720265070906e-07,
      "loss": 0.1403,
      "num_input_tokens_seen": 58734456,
      "step": 36515
    },
    {
      "epoch": 1.6532515448516263,
      "grad_norm": 1.2138476371765137,
      "learning_rate": 9.285194132456931e-07,
      "loss": 0.1645,
      "num_input_tokens_seen": 58742232,
      "step": 36520
    },
    {
      "epoch": 1.6534778967382693,
      "grad_norm": 0.30774015188217163,
      "learning_rate": 9.258705121091032e-07,
      "loss": 0.113,
      "num_input_tokens_seen": 58750360,
      "step": 36525
    },
    {
      "epoch": 1.6537042486249123,
      "grad_norm": 0.6588068604469299,
      "learning_rate": 9.232253235058136e-07,
      "loss": 0.1551,
      "num_input_tokens_seen": 58758200,
      "step": 36530
    },
    {
      "epoch": 1.6539306005115553,
      "grad_norm": 0.4320870637893677,
      "learning_rate": 9.205838478437478e-07,
      "loss": 0.1484,
      "num_input_tokens_seen": 58765880,
      "step": 36535
    },
    {
      "epoch": 1.6541569523981983,
      "grad_norm": 1.0808707475662231,
      "learning_rate": 9.179460855302524e-07,
      "loss": 0.162,
      "num_input_tokens_seen": 58773368,
      "step": 36540
    },
    {
      "epoch": 1.6543833042848413,
      "grad_norm": 0.6398759484291077,
      "learning_rate": 9.153120369721046e-07,
      "loss": 0.1329,
      "num_input_tokens_seen": 58781528,
      "step": 36545
    },
    {
      "epoch": 1.654609656171484,
      "grad_norm": 0.6750524640083313,
      "learning_rate": 9.126817025755103e-07,
      "loss": 0.1636,
      "num_input_tokens_seen": 58789592,
      "step": 36550
    },
    {
      "epoch": 1.654836008058127,
      "grad_norm": 0.2843228876590729,
      "learning_rate": 9.100550827460947e-07,
      "loss": 0.1462,
      "num_input_tokens_seen": 58797624,
      "step": 36555
    },
    {
      "epoch": 1.65506235994477,
      "grad_norm": 0.34493201971054077,
      "learning_rate": 9.0743217788892e-07,
      "loss": 0.1389,
      "num_input_tokens_seen": 58805720,
      "step": 36560
    },
    {
      "epoch": 1.655288711831413,
      "grad_norm": 0.6130955219268799,
      "learning_rate": 9.048129884084683e-07,
      "loss": 0.0975,
      "num_input_tokens_seen": 58813432,
      "step": 36565
    },
    {
      "epoch": 1.655515063718056,
      "grad_norm": 0.5570453405380249,
      "learning_rate": 9.021975147086553e-07,
      "loss": 0.1088,
      "num_input_tokens_seen": 58821016,
      "step": 36570
    },
    {
      "epoch": 1.655741415604699,
      "grad_norm": 0.6999742388725281,
      "learning_rate": 8.995857571928141e-07,
      "loss": 0.1167,
      "num_input_tokens_seen": 58828984,
      "step": 36575
    },
    {
      "epoch": 1.655967767491342,
      "grad_norm": 0.37169185280799866,
      "learning_rate": 8.969777162637139e-07,
      "loss": 0.1016,
      "num_input_tokens_seen": 58836952,
      "step": 36580
    },
    {
      "epoch": 1.656194119377985,
      "grad_norm": 0.7716563940048218,
      "learning_rate": 8.943733923235525e-07,
      "loss": 0.1433,
      "num_input_tokens_seen": 58844792,
      "step": 36585
    },
    {
      "epoch": 1.656420471264628,
      "grad_norm": 0.7992485165596008,
      "learning_rate": 8.917727857739394e-07,
      "loss": 0.1771,
      "num_input_tokens_seen": 58853080,
      "step": 36590
    },
    {
      "epoch": 1.656646823151271,
      "grad_norm": 1.2739578485488892,
      "learning_rate": 8.891758970159258e-07,
      "loss": 0.1325,
      "num_input_tokens_seen": 58861080,
      "step": 36595
    },
    {
      "epoch": 1.656873175037914,
      "grad_norm": 0.42241302132606506,
      "learning_rate": 8.86582726449986e-07,
      "loss": 0.142,
      "num_input_tokens_seen": 58869400,
      "step": 36600
    },
    {
      "epoch": 1.656873175037914,
      "eval_loss": 0.14347702264785767,
      "eval_runtime": 404.9865,
      "eval_samples_per_second": 96.969,
      "eval_steps_per_second": 24.243,
      "num_input_tokens_seen": 58869400,
      "step": 36600
    },
    {
      "epoch": 1.657099526924557,
      "grad_norm": 0.5851702690124512,
      "learning_rate": 8.839932744760165e-07,
      "loss": 0.1584,
      "num_input_tokens_seen": 58877944,
      "step": 36605
    },
    {
      "epoch": 1.6573258788112,
      "grad_norm": 0.447805792093277,
      "learning_rate": 8.814075414933482e-07,
      "loss": 0.1488,
      "num_input_tokens_seen": 58886232,
      "step": 36610
    },
    {
      "epoch": 1.657552230697843,
      "grad_norm": 0.327161967754364,
      "learning_rate": 8.788255279007257e-07,
      "loss": 0.1737,
      "num_input_tokens_seen": 58894200,
      "step": 36615
    },
    {
      "epoch": 1.657778582584486,
      "grad_norm": 1.0353260040283203,
      "learning_rate": 8.762472340963362e-07,
      "loss": 0.1195,
      "num_input_tokens_seen": 58902424,
      "step": 36620
    },
    {
      "epoch": 1.658004934471129,
      "grad_norm": 0.6363146305084229,
      "learning_rate": 8.736726604777811e-07,
      "loss": 0.1016,
      "num_input_tokens_seen": 58909816,
      "step": 36625
    },
    {
      "epoch": 1.658231286357772,
      "grad_norm": 0.5478514432907104,
      "learning_rate": 8.711018074420901e-07,
      "loss": 0.1235,
      "num_input_tokens_seen": 58917976,
      "step": 36630
    },
    {
      "epoch": 1.6584576382444147,
      "grad_norm": 0.6237629055976868,
      "learning_rate": 8.685346753857209e-07,
      "loss": 0.1176,
      "num_input_tokens_seen": 58926456,
      "step": 36635
    },
    {
      "epoch": 1.6586839901310577,
      "grad_norm": 0.41231095790863037,
      "learning_rate": 8.659712647045654e-07,
      "loss": 0.1252,
      "num_input_tokens_seen": 58935096,
      "step": 36640
    },
    {
      "epoch": 1.6589103420177007,
      "grad_norm": 0.4863142669200897,
      "learning_rate": 8.634115757939209e-07,
      "loss": 0.136,
      "num_input_tokens_seen": 58942680,
      "step": 36645
    },
    {
      "epoch": 1.6591366939043437,
      "grad_norm": 0.5571683645248413,
      "learning_rate": 8.608556090485387e-07,
      "loss": 0.1267,
      "num_input_tokens_seen": 58951352,
      "step": 36650
    },
    {
      "epoch": 1.6593630457909867,
      "grad_norm": 0.4746685326099396,
      "learning_rate": 8.583033648625671e-07,
      "loss": 0.1449,
      "num_input_tokens_seen": 58958872,
      "step": 36655
    },
    {
      "epoch": 1.6595893976776297,
      "grad_norm": 0.6016072034835815,
      "learning_rate": 8.557548436295998e-07,
      "loss": 0.1549,
      "num_input_tokens_seen": 58966808,
      "step": 36660
    },
    {
      "epoch": 1.6598157495642725,
      "grad_norm": 0.3305174708366394,
      "learning_rate": 8.532100457426556e-07,
      "loss": 0.1539,
      "num_input_tokens_seen": 58974584,
      "step": 36665
    },
    {
      "epoch": 1.6600421014509155,
      "grad_norm": 0.4790996015071869,
      "learning_rate": 8.506689715941679e-07,
      "loss": 0.1358,
      "num_input_tokens_seen": 58982776,
      "step": 36670
    },
    {
      "epoch": 1.6602684533375585,
      "grad_norm": 0.6925243139266968,
      "learning_rate": 8.481316215760011e-07,
      "loss": 0.1533,
      "num_input_tokens_seen": 58991512,
      "step": 36675
    },
    {
      "epoch": 1.6604948052242015,
      "grad_norm": 0.7395552396774292,
      "learning_rate": 8.455979960794558e-07,
      "loss": 0.1371,
      "num_input_tokens_seen": 58999480,
      "step": 36680
    },
    {
      "epoch": 1.6607211571108444,
      "grad_norm": 0.5686530470848083,
      "learning_rate": 8.430680954952364e-07,
      "loss": 0.1622,
      "num_input_tokens_seen": 59007928,
      "step": 36685
    },
    {
      "epoch": 1.6609475089974874,
      "grad_norm": 0.7247616648674011,
      "learning_rate": 8.405419202134974e-07,
      "loss": 0.1009,
      "num_input_tokens_seen": 59015608,
      "step": 36690
    },
    {
      "epoch": 1.6611738608841304,
      "grad_norm": 0.2879562973976135,
      "learning_rate": 8.380194706237993e-07,
      "loss": 0.1571,
      "num_input_tokens_seen": 59023128,
      "step": 36695
    },
    {
      "epoch": 1.6614002127707734,
      "grad_norm": 0.5188550353050232,
      "learning_rate": 8.355007471151366e-07,
      "loss": 0.1129,
      "num_input_tokens_seen": 59030776,
      "step": 36700
    },
    {
      "epoch": 1.6616265646574164,
      "grad_norm": 0.5100446939468384,
      "learning_rate": 8.329857500759292e-07,
      "loss": 0.1368,
      "num_input_tokens_seen": 59038584,
      "step": 36705
    },
    {
      "epoch": 1.6618529165440594,
      "grad_norm": 0.5898245573043823,
      "learning_rate": 8.304744798940194e-07,
      "loss": 0.1724,
      "num_input_tokens_seen": 59046648,
      "step": 36710
    },
    {
      "epoch": 1.6620792684307024,
      "grad_norm": 0.6528393626213074,
      "learning_rate": 8.279669369566756e-07,
      "loss": 0.1364,
      "num_input_tokens_seen": 59056216,
      "step": 36715
    },
    {
      "epoch": 1.6623056203173454,
      "grad_norm": 0.4273417592048645,
      "learning_rate": 8.254631216505993e-07,
      "loss": 0.1691,
      "num_input_tokens_seen": 59064248,
      "step": 36720
    },
    {
      "epoch": 1.6625319722039884,
      "grad_norm": 1.01671302318573,
      "learning_rate": 8.229630343619038e-07,
      "loss": 0.1623,
      "num_input_tokens_seen": 59071992,
      "step": 36725
    },
    {
      "epoch": 1.6627583240906314,
      "grad_norm": 0.2602728605270386,
      "learning_rate": 8.204666754761392e-07,
      "loss": 0.0956,
      "num_input_tokens_seen": 59080728,
      "step": 36730
    },
    {
      "epoch": 1.6629846759772744,
      "grad_norm": 0.36715370416641235,
      "learning_rate": 8.179740453782669e-07,
      "loss": 0.142,
      "num_input_tokens_seen": 59088024,
      "step": 36735
    },
    {
      "epoch": 1.6632110278639174,
      "grad_norm": 0.5464869141578674,
      "learning_rate": 8.154851444526907e-07,
      "loss": 0.1288,
      "num_input_tokens_seen": 59096056,
      "step": 36740
    },
    {
      "epoch": 1.6634373797505604,
      "grad_norm": 0.7182846069335938,
      "learning_rate": 8.129999730832283e-07,
      "loss": 0.1368,
      "num_input_tokens_seen": 59104088,
      "step": 36745
    },
    {
      "epoch": 1.6636637316372032,
      "grad_norm": 1.3850176334381104,
      "learning_rate": 8.105185316531178e-07,
      "loss": 0.1436,
      "num_input_tokens_seen": 59111864,
      "step": 36750
    },
    {
      "epoch": 1.6638900835238462,
      "grad_norm": 0.42046311497688293,
      "learning_rate": 8.08040820545039e-07,
      "loss": 0.1467,
      "num_input_tokens_seen": 59119448,
      "step": 36755
    },
    {
      "epoch": 1.6641164354104891,
      "grad_norm": 0.8302428126335144,
      "learning_rate": 8.055668401410782e-07,
      "loss": 0.1766,
      "num_input_tokens_seen": 59127384,
      "step": 36760
    },
    {
      "epoch": 1.6643427872971321,
      "grad_norm": 0.3655863106250763,
      "learning_rate": 8.030965908227578e-07,
      "loss": 0.1293,
      "num_input_tokens_seen": 59135352,
      "step": 36765
    },
    {
      "epoch": 1.6645691391837751,
      "grad_norm": 0.9900097846984863,
      "learning_rate": 8.006300729710203e-07,
      "loss": 0.1635,
      "num_input_tokens_seen": 59143832,
      "step": 36770
    },
    {
      "epoch": 1.664795491070418,
      "grad_norm": 0.8339362740516663,
      "learning_rate": 7.981672869662337e-07,
      "loss": 0.1224,
      "num_input_tokens_seen": 59151480,
      "step": 36775
    },
    {
      "epoch": 1.665021842957061,
      "grad_norm": 0.5489417910575867,
      "learning_rate": 7.957082331881888e-07,
      "loss": 0.1639,
      "num_input_tokens_seen": 59159448,
      "step": 36780
    },
    {
      "epoch": 1.665248194843704,
      "grad_norm": 0.7483280301094055,
      "learning_rate": 7.932529120161069e-07,
      "loss": 0.1471,
      "num_input_tokens_seen": 59167256,
      "step": 36785
    },
    {
      "epoch": 1.665474546730347,
      "grad_norm": 0.7490714192390442,
      "learning_rate": 7.908013238286243e-07,
      "loss": 0.1627,
      "num_input_tokens_seen": 59175768,
      "step": 36790
    },
    {
      "epoch": 1.66570089861699,
      "grad_norm": 0.5229106545448303,
      "learning_rate": 7.883534690038136e-07,
      "loss": 0.1295,
      "num_input_tokens_seen": 59184376,
      "step": 36795
    },
    {
      "epoch": 1.6659272505036329,
      "grad_norm": 0.57115638256073,
      "learning_rate": 7.859093479191559e-07,
      "loss": 0.1227,
      "num_input_tokens_seen": 59192472,
      "step": 36800
    },
    {
      "epoch": 1.6659272505036329,
      "eval_loss": 0.14372727274894714,
      "eval_runtime": 404.5331,
      "eval_samples_per_second": 97.077,
      "eval_steps_per_second": 24.27,
      "num_input_tokens_seen": 59192472,
      "step": 36800
    },
    {
      "epoch": 1.6661536023902759,
      "grad_norm": 0.5668554306030273,
      "learning_rate": 7.834689609515722e-07,
      "loss": 0.1042,
      "num_input_tokens_seen": 59200344,
      "step": 36805
    },
    {
      "epoch": 1.6663799542769189,
      "grad_norm": 0.7530050277709961,
      "learning_rate": 7.810323084774002e-07,
      "loss": 0.1189,
      "num_input_tokens_seen": 59208664,
      "step": 36810
    },
    {
      "epoch": 1.6666063061635619,
      "grad_norm": 0.8322173953056335,
      "learning_rate": 7.785993908723976e-07,
      "loss": 0.1456,
      "num_input_tokens_seen": 59217400,
      "step": 36815
    },
    {
      "epoch": 1.6668326580502049,
      "grad_norm": 0.4873676002025604,
      "learning_rate": 7.761702085117534e-07,
      "loss": 0.1969,
      "num_input_tokens_seen": 59225368,
      "step": 36820
    },
    {
      "epoch": 1.6670590099368479,
      "grad_norm": 0.48061755299568176,
      "learning_rate": 7.737447617700844e-07,
      "loss": 0.1342,
      "num_input_tokens_seen": 59233912,
      "step": 36825
    },
    {
      "epoch": 1.6672853618234909,
      "grad_norm": 0.7395520210266113,
      "learning_rate": 7.713230510214136e-07,
      "loss": 0.1502,
      "num_input_tokens_seen": 59241912,
      "step": 36830
    },
    {
      "epoch": 1.6675117137101338,
      "grad_norm": 0.7110515832901001,
      "learning_rate": 7.689050766392092e-07,
      "loss": 0.1317,
      "num_input_tokens_seen": 59250840,
      "step": 36835
    },
    {
      "epoch": 1.6677380655967768,
      "grad_norm": 0.9215695261955261,
      "learning_rate": 7.664908389963477e-07,
      "loss": 0.1541,
      "num_input_tokens_seen": 59258648,
      "step": 36840
    },
    {
      "epoch": 1.6679644174834198,
      "grad_norm": 0.3045901954174042,
      "learning_rate": 7.64080338465134e-07,
      "loss": 0.1662,
      "num_input_tokens_seen": 59267256,
      "step": 36845
    },
    {
      "epoch": 1.6681907693700628,
      "grad_norm": 0.6629606485366821,
      "learning_rate": 7.616735754173043e-07,
      "loss": 0.1462,
      "num_input_tokens_seen": 59275288,
      "step": 36850
    },
    {
      "epoch": 1.6684171212567058,
      "grad_norm": 0.6731159687042236,
      "learning_rate": 7.592705502240005e-07,
      "loss": 0.1408,
      "num_input_tokens_seen": 59283992,
      "step": 36855
    },
    {
      "epoch": 1.6686434731433486,
      "grad_norm": 0.930666983127594,
      "learning_rate": 7.568712632558095e-07,
      "loss": 0.159,
      "num_input_tokens_seen": 59292152,
      "step": 36860
    },
    {
      "epoch": 1.6688698250299916,
      "grad_norm": 0.6216838359832764,
      "learning_rate": 7.544757148827297e-07,
      "loss": 0.1341,
      "num_input_tokens_seen": 59300344,
      "step": 36865
    },
    {
      "epoch": 1.6690961769166346,
      "grad_norm": 0.3242719769477844,
      "learning_rate": 7.520839054741797e-07,
      "loss": 0.1196,
      "num_input_tokens_seen": 59308184,
      "step": 36870
    },
    {
      "epoch": 1.6693225288032776,
      "grad_norm": 0.8491654396057129,
      "learning_rate": 7.496958353990113e-07,
      "loss": 0.1182,
      "num_input_tokens_seen": 59316632,
      "step": 36875
    },
    {
      "epoch": 1.6695488806899206,
      "grad_norm": 0.4849887490272522,
      "learning_rate": 7.473115050254941e-07,
      "loss": 0.171,
      "num_input_tokens_seen": 59324728,
      "step": 36880
    },
    {
      "epoch": 1.6697752325765636,
      "grad_norm": 0.5405667424201965,
      "learning_rate": 7.449309147213173e-07,
      "loss": 0.1116,
      "num_input_tokens_seen": 59332120,
      "step": 36885
    },
    {
      "epoch": 1.6700015844632063,
      "grad_norm": 0.30301228165626526,
      "learning_rate": 7.425540648536067e-07,
      "loss": 0.1158,
      "num_input_tokens_seen": 59340312,
      "step": 36890
    },
    {
      "epoch": 1.6702279363498493,
      "grad_norm": 0.5097130537033081,
      "learning_rate": 7.40180955788894e-07,
      "loss": 0.1719,
      "num_input_tokens_seen": 59348472,
      "step": 36895
    },
    {
      "epoch": 1.6704542882364923,
      "grad_norm": 0.881513237953186,
      "learning_rate": 7.378115878931474e-07,
      "loss": 0.132,
      "num_input_tokens_seen": 59355896,
      "step": 36900
    },
    {
      "epoch": 1.6706806401231353,
      "grad_norm": 0.5618587136268616,
      "learning_rate": 7.354459615317527e-07,
      "loss": 0.1432,
      "num_input_tokens_seen": 59363928,
      "step": 36905
    },
    {
      "epoch": 1.6709069920097783,
      "grad_norm": 0.6481943130493164,
      "learning_rate": 7.33084077069518e-07,
      "loss": 0.137,
      "num_input_tokens_seen": 59371704,
      "step": 36910
    },
    {
      "epoch": 1.6711333438964213,
      "grad_norm": 0.8990553617477417,
      "learning_rate": 7.307259348706768e-07,
      "loss": 0.1557,
      "num_input_tokens_seen": 59380088,
      "step": 36915
    },
    {
      "epoch": 1.6713596957830643,
      "grad_norm": 0.9961262941360474,
      "learning_rate": 7.283715352988801e-07,
      "loss": 0.1645,
      "num_input_tokens_seen": 59387512,
      "step": 36920
    },
    {
      "epoch": 1.6715860476697073,
      "grad_norm": 0.6674737930297852,
      "learning_rate": 7.260208787172068e-07,
      "loss": 0.1765,
      "num_input_tokens_seen": 59395704,
      "step": 36925
    },
    {
      "epoch": 1.6718123995563503,
      "grad_norm": 1.287138819694519,
      "learning_rate": 7.23673965488167e-07,
      "loss": 0.1533,
      "num_input_tokens_seen": 59403320,
      "step": 36930
    },
    {
      "epoch": 1.6720387514429933,
      "grad_norm": 0.5938692092895508,
      "learning_rate": 7.213307959736709e-07,
      "loss": 0.1268,
      "num_input_tokens_seen": 59410808,
      "step": 36935
    },
    {
      "epoch": 1.6722651033296363,
      "grad_norm": 0.6610203981399536,
      "learning_rate": 7.189913705350715e-07,
      "loss": 0.1524,
      "num_input_tokens_seen": 59418712,
      "step": 36940
    },
    {
      "epoch": 1.6724914552162793,
      "grad_norm": 0.9685509204864502,
      "learning_rate": 7.166556895331411e-07,
      "loss": 0.1432,
      "num_input_tokens_seen": 59427160,
      "step": 36945
    },
    {
      "epoch": 1.6727178071029223,
      "grad_norm": 0.5198841691017151,
      "learning_rate": 7.143237533280639e-07,
      "loss": 0.1392,
      "num_input_tokens_seen": 59435384,
      "step": 36950
    },
    {
      "epoch": 1.6729441589895653,
      "grad_norm": 0.5138943791389465,
      "learning_rate": 7.119955622794578e-07,
      "loss": 0.1616,
      "num_input_tokens_seen": 59443448,
      "step": 36955
    },
    {
      "epoch": 1.6731705108762083,
      "grad_norm": 0.7157418131828308,
      "learning_rate": 7.096711167463577e-07,
      "loss": 0.1255,
      "num_input_tokens_seen": 59451544,
      "step": 36960
    },
    {
      "epoch": 1.6733968627628513,
      "grad_norm": 0.8777115941047668,
      "learning_rate": 7.073504170872213e-07,
      "loss": 0.1221,
      "num_input_tokens_seen": 59459768,
      "step": 36965
    },
    {
      "epoch": 1.6736232146494943,
      "grad_norm": 0.620796263217926,
      "learning_rate": 7.05033463659932e-07,
      "loss": 0.1613,
      "num_input_tokens_seen": 59467992,
      "step": 36970
    },
    {
      "epoch": 1.673849566536137,
      "grad_norm": 0.7852569222450256,
      "learning_rate": 7.027202568217928e-07,
      "loss": 0.1452,
      "num_input_tokens_seen": 59475896,
      "step": 36975
    },
    {
      "epoch": 1.67407591842278,
      "grad_norm": 0.8603322505950928,
      "learning_rate": 7.004107969295293e-07,
      "loss": 0.1358,
      "num_input_tokens_seen": 59484664,
      "step": 36980
    },
    {
      "epoch": 1.674302270309423,
      "grad_norm": 0.5564257502555847,
      "learning_rate": 6.9810508433929e-07,
      "loss": 0.1226,
      "num_input_tokens_seen": 59492344,
      "step": 36985
    },
    {
      "epoch": 1.674528622196066,
      "grad_norm": 1.0998257398605347,
      "learning_rate": 6.958031194066406e-07,
      "loss": 0.1406,
      "num_input_tokens_seen": 59500216,
      "step": 36990
    },
    {
      "epoch": 1.674754974082709,
      "grad_norm": 0.5445488691329956,
      "learning_rate": 6.935049024865776e-07,
      "loss": 0.1627,
      "num_input_tokens_seen": 59508888,
      "step": 36995
    },
    {
      "epoch": 1.674981325969352,
      "grad_norm": 0.34429988265037537,
      "learning_rate": 6.912104339335118e-07,
      "loss": 0.138,
      "num_input_tokens_seen": 59516888,
      "step": 37000
    },
    {
      "epoch": 1.674981325969352,
      "eval_loss": 0.14368689060211182,
      "eval_runtime": 404.1183,
      "eval_samples_per_second": 97.177,
      "eval_steps_per_second": 24.295,
      "num_input_tokens_seen": 59516888,
      "step": 37000
    },
    {
      "epoch": 1.6752076778559948,
      "grad_norm": 0.6831070780754089,
      "learning_rate": 6.889197141012799e-07,
      "loss": 0.1408,
      "num_input_tokens_seen": 59524984,
      "step": 37005
    },
    {
      "epoch": 1.6754340297426378,
      "grad_norm": 0.9257062673568726,
      "learning_rate": 6.866327433431435e-07,
      "loss": 0.1338,
      "num_input_tokens_seen": 59532984,
      "step": 37010
    },
    {
      "epoch": 1.6756603816292808,
      "grad_norm": 0.6889376044273376,
      "learning_rate": 6.843495220117735e-07,
      "loss": 0.1288,
      "num_input_tokens_seen": 59540984,
      "step": 37015
    },
    {
      "epoch": 1.6758867335159238,
      "grad_norm": 0.7547150254249573,
      "learning_rate": 6.820700504592798e-07,
      "loss": 0.1593,
      "num_input_tokens_seen": 59548632,
      "step": 37020
    },
    {
      "epoch": 1.6761130854025668,
      "grad_norm": 0.6185383796691895,
      "learning_rate": 6.797943290371839e-07,
      "loss": 0.1284,
      "num_input_tokens_seen": 59556824,
      "step": 37025
    },
    {
      "epoch": 1.6763394372892098,
      "grad_norm": 0.3977566063404083,
      "learning_rate": 6.775223580964274e-07,
      "loss": 0.1452,
      "num_input_tokens_seen": 59564536,
      "step": 37030
    },
    {
      "epoch": 1.6765657891758527,
      "grad_norm": 0.436259925365448,
      "learning_rate": 6.7525413798738e-07,
      "loss": 0.1419,
      "num_input_tokens_seen": 59572600,
      "step": 37035
    },
    {
      "epoch": 1.6767921410624957,
      "grad_norm": 0.5832049250602722,
      "learning_rate": 6.729896690598259e-07,
      "loss": 0.1071,
      "num_input_tokens_seen": 59580696,
      "step": 37040
    },
    {
      "epoch": 1.6770184929491387,
      "grad_norm": 0.45454713702201843,
      "learning_rate": 6.707289516629772e-07,
      "loss": 0.1169,
      "num_input_tokens_seen": 59588632,
      "step": 37045
    },
    {
      "epoch": 1.6772448448357817,
      "grad_norm": 0.8131635785102844,
      "learning_rate": 6.684719861454692e-07,
      "loss": 0.1215,
      "num_input_tokens_seen": 59596728,
      "step": 37050
    },
    {
      "epoch": 1.6774711967224247,
      "grad_norm": 0.8091123104095459,
      "learning_rate": 6.662187728553481e-07,
      "loss": 0.1641,
      "num_input_tokens_seen": 59604760,
      "step": 37055
    },
    {
      "epoch": 1.6776975486090677,
      "grad_norm": 0.6031889915466309,
      "learning_rate": 6.639693121400892e-07,
      "loss": 0.1674,
      "num_input_tokens_seen": 59614168,
      "step": 37060
    },
    {
      "epoch": 1.6779239004957107,
      "grad_norm": 1.5250623226165771,
      "learning_rate": 6.617236043465868e-07,
      "loss": 0.1801,
      "num_input_tokens_seen": 59621944,
      "step": 37065
    },
    {
      "epoch": 1.6781502523823537,
      "grad_norm": 1.167729377746582,
      "learning_rate": 6.594816498211587e-07,
      "loss": 0.1375,
      "num_input_tokens_seen": 59630424,
      "step": 37070
    },
    {
      "epoch": 1.6783766042689967,
      "grad_norm": 0.860626220703125,
      "learning_rate": 6.572434489095447e-07,
      "loss": 0.1358,
      "num_input_tokens_seen": 59638200,
      "step": 37075
    },
    {
      "epoch": 1.6786029561556397,
      "grad_norm": 0.7235546708106995,
      "learning_rate": 6.550090019568994e-07,
      "loss": 0.1439,
      "num_input_tokens_seen": 59646712,
      "step": 37080
    },
    {
      "epoch": 1.6788293080422827,
      "grad_norm": 0.4430946111679077,
      "learning_rate": 6.527783093078027e-07,
      "loss": 0.1262,
      "num_input_tokens_seen": 59654936,
      "step": 37085
    },
    {
      "epoch": 1.6790556599289255,
      "grad_norm": 1.1811715364456177,
      "learning_rate": 6.5055137130626e-07,
      "loss": 0.1877,
      "num_input_tokens_seen": 59663000,
      "step": 37090
    },
    {
      "epoch": 1.6792820118155685,
      "grad_norm": 0.6624345183372498,
      "learning_rate": 6.483281882956854e-07,
      "loss": 0.1199,
      "num_input_tokens_seen": 59670936,
      "step": 37095
    },
    {
      "epoch": 1.6795083637022115,
      "grad_norm": 0.6282939314842224,
      "learning_rate": 6.461087606189298e-07,
      "loss": 0.1218,
      "num_input_tokens_seen": 59679256,
      "step": 37100
    },
    {
      "epoch": 1.6797347155888545,
      "grad_norm": 0.451428085565567,
      "learning_rate": 6.438930886182554e-07,
      "loss": 0.1279,
      "num_input_tokens_seen": 59686680,
      "step": 37105
    },
    {
      "epoch": 1.6799610674754974,
      "grad_norm": 0.6538208723068237,
      "learning_rate": 6.416811726353417e-07,
      "loss": 0.1229,
      "num_input_tokens_seen": 59695448,
      "step": 37110
    },
    {
      "epoch": 1.6801874193621402,
      "grad_norm": 0.7348612546920776,
      "learning_rate": 6.394730130112991e-07,
      "loss": 0.142,
      "num_input_tokens_seen": 59703096,
      "step": 37115
    },
    {
      "epoch": 1.6804137712487832,
      "grad_norm": 0.5351006984710693,
      "learning_rate": 6.372686100866471e-07,
      "loss": 0.1455,
      "num_input_tokens_seen": 59711352,
      "step": 37120
    },
    {
      "epoch": 1.6806401231354262,
      "grad_norm": 0.6280522346496582,
      "learning_rate": 6.350679642013413e-07,
      "loss": 0.1429,
      "num_input_tokens_seen": 59719672,
      "step": 37125
    },
    {
      "epoch": 1.6808664750220692,
      "grad_norm": 0.48268815875053406,
      "learning_rate": 6.328710756947437e-07,
      "loss": 0.133,
      "num_input_tokens_seen": 59727896,
      "step": 37130
    },
    {
      "epoch": 1.6810928269087122,
      "grad_norm": 0.6097744107246399,
      "learning_rate": 6.306779449056416e-07,
      "loss": 0.1752,
      "num_input_tokens_seen": 59736024,
      "step": 37135
    },
    {
      "epoch": 1.6813191787953552,
      "grad_norm": 0.6385010480880737,
      "learning_rate": 6.284885721722422e-07,
      "loss": 0.1089,
      "num_input_tokens_seen": 59744184,
      "step": 37140
    },
    {
      "epoch": 1.6815455306819982,
      "grad_norm": 0.5359551310539246,
      "learning_rate": 6.26302957832181e-07,
      "loss": 0.1435,
      "num_input_tokens_seen": 59751736,
      "step": 37145
    },
    {
      "epoch": 1.6817718825686412,
      "grad_norm": 0.43383491039276123,
      "learning_rate": 6.241211022224997e-07,
      "loss": 0.1051,
      "num_input_tokens_seen": 59759832,
      "step": 37150
    },
    {
      "epoch": 1.6819982344552842,
      "grad_norm": 0.4134785532951355,
      "learning_rate": 6.219430056796732e-07,
      "loss": 0.1246,
      "num_input_tokens_seen": 59767672,
      "step": 37155
    },
    {
      "epoch": 1.6822245863419272,
      "grad_norm": 1.1878635883331299,
      "learning_rate": 6.19768668539586e-07,
      "loss": 0.1183,
      "num_input_tokens_seen": 59775608,
      "step": 37160
    },
    {
      "epoch": 1.6824509382285702,
      "grad_norm": 0.5176814794540405,
      "learning_rate": 6.175980911375528e-07,
      "loss": 0.1244,
      "num_input_tokens_seen": 59783288,
      "step": 37165
    },
    {
      "epoch": 1.6826772901152132,
      "grad_norm": 1.0017064809799194,
      "learning_rate": 6.154312738083034e-07,
      "loss": 0.1331,
      "num_input_tokens_seen": 59791640,
      "step": 37170
    },
    {
      "epoch": 1.6829036420018562,
      "grad_norm": 0.3537566065788269,
      "learning_rate": 6.132682168859843e-07,
      "loss": 0.1022,
      "num_input_tokens_seen": 59799736,
      "step": 37175
    },
    {
      "epoch": 1.6831299938884992,
      "grad_norm": 0.8830602765083313,
      "learning_rate": 6.111089207041704e-07,
      "loss": 0.1312,
      "num_input_tokens_seen": 59808088,
      "step": 37180
    },
    {
      "epoch": 1.6833563457751421,
      "grad_norm": 0.778089702129364,
      "learning_rate": 6.089533855958507e-07,
      "loss": 0.1342,
      "num_input_tokens_seen": 59816088,
      "step": 37185
    },
    {
      "epoch": 1.6835826976617851,
      "grad_norm": 0.5357309579849243,
      "learning_rate": 6.068016118934372e-07,
      "loss": 0.1437,
      "num_input_tokens_seen": 59824664,
      "step": 37190
    },
    {
      "epoch": 1.6838090495484281,
      "grad_norm": 0.5275630950927734,
      "learning_rate": 6.04653599928759e-07,
      "loss": 0.1426,
      "num_input_tokens_seen": 59832216,
      "step": 37195
    },
    {
      "epoch": 1.684035401435071,
      "grad_norm": 0.5631372332572937,
      "learning_rate": 6.025093500330675e-07,
      "loss": 0.1185,
      "num_input_tokens_seen": 59840056,
      "step": 37200
    },
    {
      "epoch": 1.684035401435071,
      "eval_loss": 0.14345553517341614,
      "eval_runtime": 404.7804,
      "eval_samples_per_second": 97.018,
      "eval_steps_per_second": 24.255,
      "num_input_tokens_seen": 59840056,
      "step": 37200
    },
    {
      "epoch": 1.684261753321714,
      "grad_norm": 0.7077875137329102,
      "learning_rate": 6.003688625370291e-07,
      "loss": 0.1678,
      "num_input_tokens_seen": 59848248,
      "step": 37205
    },
    {
      "epoch": 1.684488105208357,
      "grad_norm": 0.8922768235206604,
      "learning_rate": 5.982321377707406e-07,
      "loss": 0.1481,
      "num_input_tokens_seen": 59857176,
      "step": 37210
    },
    {
      "epoch": 1.684714457095,
      "grad_norm": 0.794424831867218,
      "learning_rate": 5.96099176063708e-07,
      "loss": 0.1581,
      "num_input_tokens_seen": 59864440,
      "step": 37215
    },
    {
      "epoch": 1.684940808981643,
      "grad_norm": 0.6486391425132751,
      "learning_rate": 5.93969977744857e-07,
      "loss": 0.1139,
      "num_input_tokens_seen": 59872056,
      "step": 37220
    },
    {
      "epoch": 1.6851671608682859,
      "grad_norm": 0.6191171407699585,
      "learning_rate": 5.918445431425445e-07,
      "loss": 0.138,
      "num_input_tokens_seen": 59880056,
      "step": 37225
    },
    {
      "epoch": 1.6853935127549287,
      "grad_norm": 1.2645936012268066,
      "learning_rate": 5.897228725845333e-07,
      "loss": 0.1766,
      "num_input_tokens_seen": 59888248,
      "step": 37230
    },
    {
      "epoch": 1.6856198646415717,
      "grad_norm": 1.1923766136169434,
      "learning_rate": 5.876049663980171e-07,
      "loss": 0.144,
      "num_input_tokens_seen": 59896504,
      "step": 37235
    },
    {
      "epoch": 1.6858462165282146,
      "grad_norm": 1.0130664110183716,
      "learning_rate": 5.854908249095959e-07,
      "loss": 0.1304,
      "num_input_tokens_seen": 59904760,
      "step": 37240
    },
    {
      "epoch": 1.6860725684148576,
      "grad_norm": 0.7332537174224854,
      "learning_rate": 5.833804484453031e-07,
      "loss": 0.1885,
      "num_input_tokens_seen": 59913272,
      "step": 37245
    },
    {
      "epoch": 1.6862989203015006,
      "grad_norm": 0.29598844051361084,
      "learning_rate": 5.81273837330587e-07,
      "loss": 0.1024,
      "num_input_tokens_seen": 59921080,
      "step": 37250
    },
    {
      "epoch": 1.6865252721881436,
      "grad_norm": 0.5823028087615967,
      "learning_rate": 5.791709918903071e-07,
      "loss": 0.1931,
      "num_input_tokens_seen": 59929336,
      "step": 37255
    },
    {
      "epoch": 1.6867516240747866,
      "grad_norm": 0.9481898546218872,
      "learning_rate": 5.770719124487483e-07,
      "loss": 0.1454,
      "num_input_tokens_seen": 59937656,
      "step": 37260
    },
    {
      "epoch": 1.6869779759614296,
      "grad_norm": 0.697806179523468,
      "learning_rate": 5.749765993296241e-07,
      "loss": 0.1402,
      "num_input_tokens_seen": 59947032,
      "step": 37265
    },
    {
      "epoch": 1.6872043278480726,
      "grad_norm": 0.8358617424964905,
      "learning_rate": 5.728850528560509e-07,
      "loss": 0.162,
      "num_input_tokens_seen": 59955224,
      "step": 37270
    },
    {
      "epoch": 1.6874306797347156,
      "grad_norm": 0.4221167266368866,
      "learning_rate": 5.707972733505707e-07,
      "loss": 0.1703,
      "num_input_tokens_seen": 59963704,
      "step": 37275
    },
    {
      "epoch": 1.6876570316213586,
      "grad_norm": 1.0691577196121216,
      "learning_rate": 5.687132611351509e-07,
      "loss": 0.1098,
      "num_input_tokens_seen": 59971640,
      "step": 37280
    },
    {
      "epoch": 1.6878833835080016,
      "grad_norm": 0.6738375425338745,
      "learning_rate": 5.666330165311651e-07,
      "loss": 0.1332,
      "num_input_tokens_seen": 59979480,
      "step": 37285
    },
    {
      "epoch": 1.6881097353946446,
      "grad_norm": 0.7272816300392151,
      "learning_rate": 5.645565398594204e-07,
      "loss": 0.1249,
      "num_input_tokens_seen": 59987800,
      "step": 37290
    },
    {
      "epoch": 1.6883360872812876,
      "grad_norm": 0.4851168990135193,
      "learning_rate": 5.624838314401304e-07,
      "loss": 0.1012,
      "num_input_tokens_seen": 59996216,
      "step": 37295
    },
    {
      "epoch": 1.6885624391679306,
      "grad_norm": 0.3820684552192688,
      "learning_rate": 5.604148915929336e-07,
      "loss": 0.1423,
      "num_input_tokens_seen": 60003608,
      "step": 37300
    },
    {
      "epoch": 1.6887887910545736,
      "grad_norm": 0.8390443325042725,
      "learning_rate": 5.583497206368887e-07,
      "loss": 0.1935,
      "num_input_tokens_seen": 60011352,
      "step": 37305
    },
    {
      "epoch": 1.6890151429412166,
      "grad_norm": 0.2542877793312073,
      "learning_rate": 5.562883188904688e-07,
      "loss": 0.1249,
      "num_input_tokens_seen": 60020056,
      "step": 37310
    },
    {
      "epoch": 1.6892414948278593,
      "grad_norm": 0.8331578969955444,
      "learning_rate": 5.542306866715724e-07,
      "loss": 0.1023,
      "num_input_tokens_seen": 60028696,
      "step": 37315
    },
    {
      "epoch": 1.6894678467145023,
      "grad_norm": 0.63265061378479,
      "learning_rate": 5.52176824297504e-07,
      "loss": 0.1154,
      "num_input_tokens_seen": 60037688,
      "step": 37320
    },
    {
      "epoch": 1.6896941986011453,
      "grad_norm": 0.29997381567955017,
      "learning_rate": 5.501267320850018e-07,
      "loss": 0.1315,
      "num_input_tokens_seen": 60046328,
      "step": 37325
    },
    {
      "epoch": 1.6899205504877883,
      "grad_norm": 0.9246199727058411,
      "learning_rate": 5.480804103502157e-07,
      "loss": 0.1724,
      "num_input_tokens_seen": 60054616,
      "step": 37330
    },
    {
      "epoch": 1.6901469023744313,
      "grad_norm": 0.30936166644096375,
      "learning_rate": 5.460378594087101e-07,
      "loss": 0.144,
      "num_input_tokens_seen": 60062104,
      "step": 37335
    },
    {
      "epoch": 1.6903732542610743,
      "grad_norm": 0.37085339426994324,
      "learning_rate": 5.439990795754773e-07,
      "loss": 0.1336,
      "num_input_tokens_seen": 60070520,
      "step": 37340
    },
    {
      "epoch": 1.690599606147717,
      "grad_norm": 0.51091468334198,
      "learning_rate": 5.419640711649188e-07,
      "loss": 0.1096,
      "num_input_tokens_seen": 60078904,
      "step": 37345
    },
    {
      "epoch": 1.69082595803436,
      "grad_norm": 0.43782830238342285,
      "learning_rate": 5.399328344908583e-07,
      "loss": 0.1229,
      "num_input_tokens_seen": 60087224,
      "step": 37350
    },
    {
      "epoch": 1.691052309921003,
      "grad_norm": 1.201797604560852,
      "learning_rate": 5.379053698665399e-07,
      "loss": 0.1447,
      "num_input_tokens_seen": 60094936,
      "step": 37355
    },
    {
      "epoch": 1.691278661807646,
      "grad_norm": 0.828157901763916,
      "learning_rate": 5.358816776046216e-07,
      "loss": 0.1904,
      "num_input_tokens_seen": 60102488,
      "step": 37360
    },
    {
      "epoch": 1.691505013694289,
      "grad_norm": 0.37303122878074646,
      "learning_rate": 5.338617580171817e-07,
      "loss": 0.1376,
      "num_input_tokens_seen": 60110008,
      "step": 37365
    },
    {
      "epoch": 1.691731365580932,
      "grad_norm": 0.4139183759689331,
      "learning_rate": 5.318456114157239e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 60118200,
      "step": 37370
    },
    {
      "epoch": 1.691957717467575,
      "grad_norm": 0.3107753098011017,
      "learning_rate": 5.298332381111576e-07,
      "loss": 0.1265,
      "num_input_tokens_seen": 60126040,
      "step": 37375
    },
    {
      "epoch": 1.692184069354218,
      "grad_norm": 0.3511444330215454,
      "learning_rate": 5.27824638413818e-07,
      "loss": 0.2072,
      "num_input_tokens_seen": 60134328,
      "step": 37380
    },
    {
      "epoch": 1.692410421240861,
      "grad_norm": 0.7789907455444336,
      "learning_rate": 5.258198126334546e-07,
      "loss": 0.1585,
      "num_input_tokens_seen": 60142424,
      "step": 37385
    },
    {
      "epoch": 1.692636773127504,
      "grad_norm": 1.4825489521026611,
      "learning_rate": 5.238187610792367e-07,
      "loss": 0.1543,
      "num_input_tokens_seen": 60150584,
      "step": 37390
    },
    {
      "epoch": 1.692863125014147,
      "grad_norm": 0.5464069843292236,
      "learning_rate": 5.218214840597563e-07,
      "loss": 0.1554,
      "num_input_tokens_seen": 60158904,
      "step": 37395
    },
    {
      "epoch": 1.69308947690079,
      "grad_norm": 0.2553105652332306,
      "learning_rate": 5.198279818830115e-07,
      "loss": 0.0866,
      "num_input_tokens_seen": 60166712,
      "step": 37400
    },
    {
      "epoch": 1.69308947690079,
      "eval_loss": 0.14349201321601868,
      "eval_runtime": 405.3777,
      "eval_samples_per_second": 96.875,
      "eval_steps_per_second": 24.219,
      "num_input_tokens_seen": 60166712,
      "step": 37400
    },
    {
      "epoch": 1.693315828787433,
      "grad_norm": 0.6120836734771729,
      "learning_rate": 5.178382548564287e-07,
      "loss": 0.1421,
      "num_input_tokens_seen": 60174712,
      "step": 37405
    },
    {
      "epoch": 1.693542180674076,
      "grad_norm": 0.5096301436424255,
      "learning_rate": 5.15852303286854e-07,
      "loss": 0.1384,
      "num_input_tokens_seen": 60182584,
      "step": 37410
    },
    {
      "epoch": 1.693768532560719,
      "grad_norm": 0.5529988408088684,
      "learning_rate": 5.138701274805396e-07,
      "loss": 0.1338,
      "num_input_tokens_seen": 60190456,
      "step": 37415
    },
    {
      "epoch": 1.693994884447362,
      "grad_norm": 1.3134254217147827,
      "learning_rate": 5.118917277431606e-07,
      "loss": 0.1325,
      "num_input_tokens_seen": 60198008,
      "step": 37420
    },
    {
      "epoch": 1.694221236334005,
      "grad_norm": 0.8619205355644226,
      "learning_rate": 5.099171043798145e-07,
      "loss": 0.17,
      "num_input_tokens_seen": 60206488,
      "step": 37425
    },
    {
      "epoch": 1.6944475882206478,
      "grad_norm": 0.7488291263580322,
      "learning_rate": 5.079462576950133e-07,
      "loss": 0.1085,
      "num_input_tokens_seen": 60214872,
      "step": 37430
    },
    {
      "epoch": 1.6946739401072908,
      "grad_norm": 0.5078256130218506,
      "learning_rate": 5.059791879926862e-07,
      "loss": 0.1256,
      "num_input_tokens_seen": 60223064,
      "step": 37435
    },
    {
      "epoch": 1.6949002919939338,
      "grad_norm": 0.535878598690033,
      "learning_rate": 5.040158955761793e-07,
      "loss": 0.1306,
      "num_input_tokens_seen": 60232632,
      "step": 37440
    },
    {
      "epoch": 1.6951266438805768,
      "grad_norm": 0.5326722264289856,
      "learning_rate": 5.020563807482559e-07,
      "loss": 0.1295,
      "num_input_tokens_seen": 60240312,
      "step": 37445
    },
    {
      "epoch": 1.6953529957672198,
      "grad_norm": 0.5684192180633545,
      "learning_rate": 5.001006438110995e-07,
      "loss": 0.1277,
      "num_input_tokens_seen": 60249080,
      "step": 37450
    },
    {
      "epoch": 1.6955793476538625,
      "grad_norm": 0.8827987313270569,
      "learning_rate": 4.981486850663075e-07,
      "loss": 0.1613,
      "num_input_tokens_seen": 60257080,
      "step": 37455
    },
    {
      "epoch": 1.6958056995405055,
      "grad_norm": 0.4734792411327362,
      "learning_rate": 4.962005048149005e-07,
      "loss": 0.1781,
      "num_input_tokens_seen": 60264696,
      "step": 37460
    },
    {
      "epoch": 1.6960320514271485,
      "grad_norm": 0.6952320337295532,
      "learning_rate": 4.942561033573073e-07,
      "loss": 0.1492,
      "num_input_tokens_seen": 60272344,
      "step": 37465
    },
    {
      "epoch": 1.6962584033137915,
      "grad_norm": 0.6334895491600037,
      "learning_rate": 4.923154809933827e-07,
      "loss": 0.1221,
      "num_input_tokens_seen": 60280440,
      "step": 37470
    },
    {
      "epoch": 1.6964847552004345,
      "grad_norm": 0.7061992287635803,
      "learning_rate": 4.903786380223957e-07,
      "loss": 0.1681,
      "num_input_tokens_seen": 60288152,
      "step": 37475
    },
    {
      "epoch": 1.6967111070870775,
      "grad_norm": 0.8958168029785156,
      "learning_rate": 4.884455747430266e-07,
      "loss": 0.1519,
      "num_input_tokens_seen": 60296184,
      "step": 37480
    },
    {
      "epoch": 1.6969374589737205,
      "grad_norm": 0.3298494219779968,
      "learning_rate": 4.865162914533816e-07,
      "loss": 0.1256,
      "num_input_tokens_seen": 60304536,
      "step": 37485
    },
    {
      "epoch": 1.6971638108603635,
      "grad_norm": 1.0682170391082764,
      "learning_rate": 4.845907884509809e-07,
      "loss": 0.1554,
      "num_input_tokens_seen": 60311960,
      "step": 37490
    },
    {
      "epoch": 1.6973901627470065,
      "grad_norm": 0.817066490650177,
      "learning_rate": 4.82669066032762e-07,
      "loss": 0.1448,
      "num_input_tokens_seen": 60319960,
      "step": 37495
    },
    {
      "epoch": 1.6976165146336495,
      "grad_norm": 0.37096917629241943,
      "learning_rate": 4.807511244950768e-07,
      "loss": 0.1349,
      "num_input_tokens_seen": 60327896,
      "step": 37500
    },
    {
      "epoch": 1.6978428665202925,
      "grad_norm": 0.4117515981197357,
      "learning_rate": 4.788369641336943e-07,
      "loss": 0.1578,
      "num_input_tokens_seen": 60335352,
      "step": 37505
    },
    {
      "epoch": 1.6980692184069355,
      "grad_norm": 0.46342721581459045,
      "learning_rate": 4.769265852438032e-07,
      "loss": 0.1332,
      "num_input_tokens_seen": 60343800,
      "step": 37510
    },
    {
      "epoch": 1.6982955702935785,
      "grad_norm": 0.36646798253059387,
      "learning_rate": 4.750199881200124e-07,
      "loss": 0.1744,
      "num_input_tokens_seen": 60351160,
      "step": 37515
    },
    {
      "epoch": 1.6985219221802215,
      "grad_norm": 0.8562902212142944,
      "learning_rate": 4.7311717305633664e-07,
      "loss": 0.1716,
      "num_input_tokens_seen": 60359480,
      "step": 37520
    },
    {
      "epoch": 1.6987482740668645,
      "grad_norm": 1.1479332447052002,
      "learning_rate": 4.7121814034621623e-07,
      "loss": 0.1468,
      "num_input_tokens_seen": 60368792,
      "step": 37525
    },
    {
      "epoch": 1.6989746259535075,
      "grad_norm": 1.248091220855713,
      "learning_rate": 4.693228902825114e-07,
      "loss": 0.1246,
      "num_input_tokens_seen": 60377208,
      "step": 37530
    },
    {
      "epoch": 1.6992009778401505,
      "grad_norm": 0.6422452926635742,
      "learning_rate": 4.6743142315748277e-07,
      "loss": 0.1449,
      "num_input_tokens_seen": 60384920,
      "step": 37535
    },
    {
      "epoch": 1.6994273297267932,
      "grad_norm": 0.5392604470252991,
      "learning_rate": 4.655437392628276e-07,
      "loss": 0.1221,
      "num_input_tokens_seen": 60393272,
      "step": 37540
    },
    {
      "epoch": 1.6996536816134362,
      "grad_norm": 0.9149760007858276,
      "learning_rate": 4.636598388896463e-07,
      "loss": 0.1112,
      "num_input_tokens_seen": 60400888,
      "step": 37545
    },
    {
      "epoch": 1.6998800335000792,
      "grad_norm": 0.3099627196788788,
      "learning_rate": 4.6177972232845925e-07,
      "loss": 0.0919,
      "num_input_tokens_seen": 60408632,
      "step": 37550
    },
    {
      "epoch": 1.7001063853867222,
      "grad_norm": 0.4101646840572357,
      "learning_rate": 4.5990338986920953e-07,
      "loss": 0.1277,
      "num_input_tokens_seen": 60416344,
      "step": 37555
    },
    {
      "epoch": 1.7003327372733652,
      "grad_norm": 0.8991594314575195,
      "learning_rate": 4.5803084180124633e-07,
      "loss": 0.1502,
      "num_input_tokens_seen": 60424824,
      "step": 37560
    },
    {
      "epoch": 1.7005590891600082,
      "grad_norm": 0.42425763607025146,
      "learning_rate": 4.561620784133386e-07,
      "loss": 0.1351,
      "num_input_tokens_seen": 60432888,
      "step": 37565
    },
    {
      "epoch": 1.700785441046651,
      "grad_norm": 0.2520083785057068,
      "learning_rate": 4.5429709999367796e-07,
      "loss": 0.1024,
      "num_input_tokens_seen": 60441304,
      "step": 37570
    },
    {
      "epoch": 1.701011792933294,
      "grad_norm": 0.28595080971717834,
      "learning_rate": 4.5243590682986223e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 60448856,
      "step": 37575
    },
    {
      "epoch": 1.701238144819937,
      "grad_norm": 0.5761033296585083,
      "learning_rate": 4.5057849920891735e-07,
      "loss": 0.0918,
      "num_input_tokens_seen": 60456760,
      "step": 37580
    },
    {
      "epoch": 1.70146449670658,
      "grad_norm": 0.38145819306373596,
      "learning_rate": 4.487248774172698e-07,
      "loss": 0.1319,
      "num_input_tokens_seen": 60464632,
      "step": 37585
    },
    {
      "epoch": 1.701690848593223,
      "grad_norm": 0.4689885973930359,
      "learning_rate": 4.4687504174077965e-07,
      "loss": 0.1418,
      "num_input_tokens_seen": 60472696,
      "step": 37590
    },
    {
      "epoch": 1.701917200479866,
      "grad_norm": 0.44065314531326294,
      "learning_rate": 4.450289924647133e-07,
      "loss": 0.1275,
      "num_input_tokens_seen": 60480888,
      "step": 37595
    },
    {
      "epoch": 1.702143552366509,
      "grad_norm": 0.8499207496643066,
      "learning_rate": 4.431867298737513e-07,
      "loss": 0.1334,
      "num_input_tokens_seen": 60488984,
      "step": 37600
    },
    {
      "epoch": 1.702143552366509,
      "eval_loss": 0.14377522468566895,
      "eval_runtime": 405.0016,
      "eval_samples_per_second": 96.965,
      "eval_steps_per_second": 24.242,
      "num_input_tokens_seen": 60488984,
      "step": 37600
    },
    {
      "epoch": 1.702369904253152,
      "grad_norm": 1.1122866868972778,
      "learning_rate": 4.41348254251997e-07,
      "loss": 0.1376,
      "num_input_tokens_seen": 60496568,
      "step": 37605
    },
    {
      "epoch": 1.702596256139795,
      "grad_norm": 0.29324641823768616,
      "learning_rate": 4.395135658829652e-07,
      "loss": 0.1373,
      "num_input_tokens_seen": 60504408,
      "step": 37610
    },
    {
      "epoch": 1.702822608026438,
      "grad_norm": 0.44411489367485046,
      "learning_rate": 4.376826650495852e-07,
      "loss": 0.1408,
      "num_input_tokens_seen": 60513208,
      "step": 37615
    },
    {
      "epoch": 1.703048959913081,
      "grad_norm": 0.7821709513664246,
      "learning_rate": 4.358555520342117e-07,
      "loss": 0.1385,
      "num_input_tokens_seen": 60521272,
      "step": 37620
    },
    {
      "epoch": 1.703275311799724,
      "grad_norm": 0.6930923461914062,
      "learning_rate": 4.3403222711860257e-07,
      "loss": 0.1627,
      "num_input_tokens_seen": 60528600,
      "step": 37625
    },
    {
      "epoch": 1.703501663686367,
      "grad_norm": 0.406610906124115,
      "learning_rate": 4.3221269058394133e-07,
      "loss": 0.1483,
      "num_input_tokens_seen": 60536376,
      "step": 37630
    },
    {
      "epoch": 1.70372801557301,
      "grad_norm": 0.5847944021224976,
      "learning_rate": 4.303969427108173e-07,
      "loss": 0.0934,
      "num_input_tokens_seen": 60544536,
      "step": 37635
    },
    {
      "epoch": 1.703954367459653,
      "grad_norm": 0.6083574891090393,
      "learning_rate": 4.2858498377924825e-07,
      "loss": 0.1737,
      "num_input_tokens_seen": 60552632,
      "step": 37640
    },
    {
      "epoch": 1.704180719346296,
      "grad_norm": 0.3443274199962616,
      "learning_rate": 4.267768140686579e-07,
      "loss": 0.1361,
      "num_input_tokens_seen": 60560440,
      "step": 37645
    },
    {
      "epoch": 1.7044070712329389,
      "grad_norm": 0.7911412715911865,
      "learning_rate": 4.2497243385788975e-07,
      "loss": 0.1537,
      "num_input_tokens_seen": 60568536,
      "step": 37650
    },
    {
      "epoch": 1.7046334231195817,
      "grad_norm": 0.7571442127227783,
      "learning_rate": 4.231718434251991e-07,
      "loss": 0.157,
      "num_input_tokens_seen": 60576792,
      "step": 37655
    },
    {
      "epoch": 1.7048597750062247,
      "grad_norm": 0.6838545799255371,
      "learning_rate": 4.213750430482666e-07,
      "loss": 0.1782,
      "num_input_tokens_seen": 60585048,
      "step": 37660
    },
    {
      "epoch": 1.7050861268928676,
      "grad_norm": 0.5817843675613403,
      "learning_rate": 4.1958203300417054e-07,
      "loss": 0.1223,
      "num_input_tokens_seen": 60593880,
      "step": 37665
    },
    {
      "epoch": 1.7053124787795106,
      "grad_norm": 0.6683773994445801,
      "learning_rate": 4.177928135694259e-07,
      "loss": 0.1554,
      "num_input_tokens_seen": 60602200,
      "step": 37670
    },
    {
      "epoch": 1.7055388306661536,
      "grad_norm": 1.5068974494934082,
      "learning_rate": 4.1600738501994807e-07,
      "loss": 0.168,
      "num_input_tokens_seen": 60610424,
      "step": 37675
    },
    {
      "epoch": 1.7057651825527966,
      "grad_norm": 0.9240637421607971,
      "learning_rate": 4.1422574763107237e-07,
      "loss": 0.1438,
      "num_input_tokens_seen": 60618520,
      "step": 37680
    },
    {
      "epoch": 1.7059915344394394,
      "grad_norm": 0.6744527816772461,
      "learning_rate": 4.124479016775512e-07,
      "loss": 0.1113,
      "num_input_tokens_seen": 60626776,
      "step": 37685
    },
    {
      "epoch": 1.7062178863260824,
      "grad_norm": 0.41235339641571045,
      "learning_rate": 4.106738474335514e-07,
      "loss": 0.153,
      "num_input_tokens_seen": 60634488,
      "step": 37690
    },
    {
      "epoch": 1.7064442382127254,
      "grad_norm": 0.8057461380958557,
      "learning_rate": 4.089035851726486e-07,
      "loss": 0.1797,
      "num_input_tokens_seen": 60642136,
      "step": 37695
    },
    {
      "epoch": 1.7066705900993684,
      "grad_norm": 1.32837975025177,
      "learning_rate": 4.0713711516784937e-07,
      "loss": 0.1751,
      "num_input_tokens_seen": 60650040,
      "step": 37700
    },
    {
      "epoch": 1.7068969419860114,
      "grad_norm": 1.7817039489746094,
      "learning_rate": 4.05374437691558e-07,
      "loss": 0.1584,
      "num_input_tokens_seen": 60657752,
      "step": 37705
    },
    {
      "epoch": 1.7071232938726544,
      "grad_norm": 0.49927759170532227,
      "learning_rate": 4.036155530156044e-07,
      "loss": 0.1197,
      "num_input_tokens_seen": 60665976,
      "step": 37710
    },
    {
      "epoch": 1.7073496457592974,
      "grad_norm": 1.1849110126495361,
      "learning_rate": 4.018604614112298e-07,
      "loss": 0.1279,
      "num_input_tokens_seen": 60674712,
      "step": 37715
    },
    {
      "epoch": 1.7075759976459404,
      "grad_norm": 0.3487739562988281,
      "learning_rate": 4.0010916314908996e-07,
      "loss": 0.1536,
      "num_input_tokens_seen": 60683000,
      "step": 37720
    },
    {
      "epoch": 1.7078023495325834,
      "grad_norm": 0.8396360278129578,
      "learning_rate": 3.983616584992578e-07,
      "loss": 0.1374,
      "num_input_tokens_seen": 60690712,
      "step": 37725
    },
    {
      "epoch": 1.7080287014192264,
      "grad_norm": 0.6310739517211914,
      "learning_rate": 3.9661794773122595e-07,
      "loss": 0.1545,
      "num_input_tokens_seen": 60698520,
      "step": 37730
    },
    {
      "epoch": 1.7082550533058694,
      "grad_norm": 0.8615487813949585,
      "learning_rate": 3.9487803111388777e-07,
      "loss": 0.1231,
      "num_input_tokens_seen": 60706264,
      "step": 37735
    },
    {
      "epoch": 1.7084814051925123,
      "grad_norm": 0.5100764632225037,
      "learning_rate": 3.9314190891556747e-07,
      "loss": 0.1069,
      "num_input_tokens_seen": 60713848,
      "step": 37740
    },
    {
      "epoch": 1.7087077570791553,
      "grad_norm": 0.5243041515350342,
      "learning_rate": 3.914095814039925e-07,
      "loss": 0.1009,
      "num_input_tokens_seen": 60722200,
      "step": 37745
    },
    {
      "epoch": 1.7089341089657983,
      "grad_norm": 0.9313321709632874,
      "learning_rate": 3.896810488463104e-07,
      "loss": 0.1293,
      "num_input_tokens_seen": 60730680,
      "step": 37750
    },
    {
      "epoch": 1.7091604608524413,
      "grad_norm": 0.6015470623970032,
      "learning_rate": 3.8795631150908565e-07,
      "loss": 0.1218,
      "num_input_tokens_seen": 60738680,
      "step": 37755
    },
    {
      "epoch": 1.7093868127390843,
      "grad_norm": 0.699343740940094,
      "learning_rate": 3.862353696582888e-07,
      "loss": 0.1435,
      "num_input_tokens_seen": 60747160,
      "step": 37760
    },
    {
      "epoch": 1.7096131646257273,
      "grad_norm": 0.8000585436820984,
      "learning_rate": 3.8451822355931313e-07,
      "loss": 0.1211,
      "num_input_tokens_seen": 60755352,
      "step": 37765
    },
    {
      "epoch": 1.70983951651237,
      "grad_norm": 0.33672893047332764,
      "learning_rate": 3.82804873476969e-07,
      "loss": 0.1477,
      "num_input_tokens_seen": 60763192,
      "step": 37770
    },
    {
      "epoch": 1.710065868399013,
      "grad_norm": 0.46673986315727234,
      "learning_rate": 3.810953196754702e-07,
      "loss": 0.1586,
      "num_input_tokens_seen": 60771064,
      "step": 37775
    },
    {
      "epoch": 1.710292220285656,
      "grad_norm": 0.8373739719390869,
      "learning_rate": 3.793895624184529e-07,
      "loss": 0.1744,
      "num_input_tokens_seen": 60778840,
      "step": 37780
    },
    {
      "epoch": 1.710518572172299,
      "grad_norm": 1.0459011793136597,
      "learning_rate": 3.776876019689679e-07,
      "loss": 0.1212,
      "num_input_tokens_seen": 60786744,
      "step": 37785
    },
    {
      "epoch": 1.710744924058942,
      "grad_norm": 0.6794412136077881,
      "learning_rate": 3.7598943858947743e-07,
      "loss": 0.1463,
      "num_input_tokens_seen": 60795064,
      "step": 37790
    },
    {
      "epoch": 1.7109712759455848,
      "grad_norm": 1.065802812576294,
      "learning_rate": 3.742950725418637e-07,
      "loss": 0.1337,
      "num_input_tokens_seen": 60802776,
      "step": 37795
    },
    {
      "epoch": 1.7111976278322278,
      "grad_norm": 0.8566538095474243,
      "learning_rate": 3.726045040874093e-07,
      "loss": 0.1368,
      "num_input_tokens_seen": 60810360,
      "step": 37800
    },
    {
      "epoch": 1.7111976278322278,
      "eval_loss": 0.14336636662483215,
      "eval_runtime": 404.656,
      "eval_samples_per_second": 97.048,
      "eval_steps_per_second": 24.263,
      "num_input_tokens_seen": 60810360,
      "step": 37800
    },
    {
      "epoch": 1.7114239797188708,
      "grad_norm": 0.9385923743247986,
      "learning_rate": 3.709177334868308e-07,
      "loss": 0.1244,
      "num_input_tokens_seen": 60818392,
      "step": 37805
    },
    {
      "epoch": 1.7116503316055138,
      "grad_norm": 0.26996326446533203,
      "learning_rate": 3.692347610002478e-07,
      "loss": 0.1312,
      "num_input_tokens_seen": 60826392,
      "step": 37810
    },
    {
      "epoch": 1.7118766834921568,
      "grad_norm": 0.47610095143318176,
      "learning_rate": 3.675555868871916e-07,
      "loss": 0.1525,
      "num_input_tokens_seen": 60834424,
      "step": 37815
    },
    {
      "epoch": 1.7121030353787998,
      "grad_norm": 0.5904646515846252,
      "learning_rate": 3.658802114066162e-07,
      "loss": 0.1623,
      "num_input_tokens_seen": 60842104,
      "step": 37820
    },
    {
      "epoch": 1.7123293872654428,
      "grad_norm": 0.5762553811073303,
      "learning_rate": 3.6420863481688437e-07,
      "loss": 0.1619,
      "num_input_tokens_seen": 60850488,
      "step": 37825
    },
    {
      "epoch": 1.7125557391520858,
      "grad_norm": 0.8840634226799011,
      "learning_rate": 3.625408573757705e-07,
      "loss": 0.1736,
      "num_input_tokens_seen": 60857944,
      "step": 37830
    },
    {
      "epoch": 1.7127820910387288,
      "grad_norm": 1.052916407585144,
      "learning_rate": 3.608768793404743e-07,
      "loss": 0.1504,
      "num_input_tokens_seen": 60866072,
      "step": 37835
    },
    {
      "epoch": 1.7130084429253718,
      "grad_norm": 0.3303286135196686,
      "learning_rate": 3.592167009675934e-07,
      "loss": 0.1485,
      "num_input_tokens_seen": 60874552,
      "step": 37840
    },
    {
      "epoch": 1.7132347948120148,
      "grad_norm": 0.6481484770774841,
      "learning_rate": 3.575603225131563e-07,
      "loss": 0.1317,
      "num_input_tokens_seen": 60883032,
      "step": 37845
    },
    {
      "epoch": 1.7134611466986578,
      "grad_norm": 0.6331541538238525,
      "learning_rate": 3.55907744232592e-07,
      "loss": 0.1253,
      "num_input_tokens_seen": 60890712,
      "step": 37850
    },
    {
      "epoch": 1.7136874985853008,
      "grad_norm": 0.7675436735153198,
      "learning_rate": 3.5425896638075217e-07,
      "loss": 0.1526,
      "num_input_tokens_seen": 60899384,
      "step": 37855
    },
    {
      "epoch": 1.7139138504719438,
      "grad_norm": 0.3951790928840637,
      "learning_rate": 3.5261398921189736e-07,
      "loss": 0.1242,
      "num_input_tokens_seen": 60907480,
      "step": 37860
    },
    {
      "epoch": 1.7141402023585868,
      "grad_norm": 0.3898828327655792,
      "learning_rate": 3.509728129797024e-07,
      "loss": 0.1084,
      "num_input_tokens_seen": 60915832,
      "step": 37865
    },
    {
      "epoch": 1.7143665542452298,
      "grad_norm": 0.8449939489364624,
      "learning_rate": 3.4933543793725656e-07,
      "loss": 0.1661,
      "num_input_tokens_seen": 60923864,
      "step": 37870
    },
    {
      "epoch": 1.7145929061318728,
      "grad_norm": 0.5486172437667847,
      "learning_rate": 3.4770186433707163e-07,
      "loss": 0.1374,
      "num_input_tokens_seen": 60931256,
      "step": 37875
    },
    {
      "epoch": 1.7148192580185155,
      "grad_norm": 0.4284263253211975,
      "learning_rate": 3.4607209243105453e-07,
      "loss": 0.1309,
      "num_input_tokens_seen": 60939608,
      "step": 37880
    },
    {
      "epoch": 1.7150456099051585,
      "grad_norm": 0.6107932925224304,
      "learning_rate": 3.444461224705431e-07,
      "loss": 0.1794,
      "num_input_tokens_seen": 60947576,
      "step": 37885
    },
    {
      "epoch": 1.7152719617918015,
      "grad_norm": 0.5515244603157043,
      "learning_rate": 3.4282395470628116e-07,
      "loss": 0.1295,
      "num_input_tokens_seen": 60955416,
      "step": 37890
    },
    {
      "epoch": 1.7154983136784445,
      "grad_norm": 0.7215771675109863,
      "learning_rate": 3.4120558938842417e-07,
      "loss": 0.0895,
      "num_input_tokens_seen": 60963480,
      "step": 37895
    },
    {
      "epoch": 1.7157246655650875,
      "grad_norm": 0.8190429210662842,
      "learning_rate": 3.395910267665503e-07,
      "loss": 0.1283,
      "num_input_tokens_seen": 60971288,
      "step": 37900
    },
    {
      "epoch": 1.7159510174517305,
      "grad_norm": 0.2910557687282562,
      "learning_rate": 3.3798026708964094e-07,
      "loss": 0.1154,
      "num_input_tokens_seen": 60978840,
      "step": 37905
    },
    {
      "epoch": 1.7161773693383733,
      "grad_norm": 0.389702707529068,
      "learning_rate": 3.3637331060609456e-07,
      "loss": 0.1133,
      "num_input_tokens_seen": 60986936,
      "step": 37910
    },
    {
      "epoch": 1.7164037212250163,
      "grad_norm": 0.46161115169525146,
      "learning_rate": 3.3477015756372966e-07,
      "loss": 0.1085,
      "num_input_tokens_seen": 60994904,
      "step": 37915
    },
    {
      "epoch": 1.7166300731116593,
      "grad_norm": 0.6817752122879028,
      "learning_rate": 3.3317080820976785e-07,
      "loss": 0.1548,
      "num_input_tokens_seen": 61002936,
      "step": 37920
    },
    {
      "epoch": 1.7168564249983023,
      "grad_norm": 0.6471911668777466,
      "learning_rate": 3.315752627908508e-07,
      "loss": 0.1092,
      "num_input_tokens_seen": 61010776,
      "step": 37925
    },
    {
      "epoch": 1.7170827768849453,
      "grad_norm": 0.46620845794677734,
      "learning_rate": 3.299835215530317e-07,
      "loss": 0.1325,
      "num_input_tokens_seen": 61018872,
      "step": 37930
    },
    {
      "epoch": 1.7173091287715883,
      "grad_norm": 0.37602412700653076,
      "learning_rate": 3.2839558474177245e-07,
      "loss": 0.1322,
      "num_input_tokens_seen": 61027544,
      "step": 37935
    },
    {
      "epoch": 1.7175354806582313,
      "grad_norm": 0.8263082504272461,
      "learning_rate": 3.2681145260196056e-07,
      "loss": 0.1449,
      "num_input_tokens_seen": 61035288,
      "step": 37940
    },
    {
      "epoch": 1.7177618325448742,
      "grad_norm": 0.49722957611083984,
      "learning_rate": 3.252311253778839e-07,
      "loss": 0.139,
      "num_input_tokens_seen": 61044280,
      "step": 37945
    },
    {
      "epoch": 1.7179881844315172,
      "grad_norm": 0.8103669285774231,
      "learning_rate": 3.2365460331325034e-07,
      "loss": 0.1209,
      "num_input_tokens_seen": 61052760,
      "step": 37950
    },
    {
      "epoch": 1.7182145363181602,
      "grad_norm": 1.3928093910217285,
      "learning_rate": 3.2208188665117934e-07,
      "loss": 0.139,
      "num_input_tokens_seen": 61060632,
      "step": 37955
    },
    {
      "epoch": 1.7184408882048032,
      "grad_norm": 0.6563168168067932,
      "learning_rate": 3.205129756342018e-07,
      "loss": 0.186,
      "num_input_tokens_seen": 61068984,
      "step": 37960
    },
    {
      "epoch": 1.7186672400914462,
      "grad_norm": 0.8361153602600098,
      "learning_rate": 3.189478705042659e-07,
      "loss": 0.1615,
      "num_input_tokens_seen": 61077464,
      "step": 37965
    },
    {
      "epoch": 1.7188935919780892,
      "grad_norm": 0.5226202011108398,
      "learning_rate": 3.173865715027341e-07,
      "loss": 0.1171,
      "num_input_tokens_seen": 61085048,
      "step": 37970
    },
    {
      "epoch": 1.7191199438647322,
      "grad_norm": 0.4111439883708954,
      "learning_rate": 3.158290788703694e-07,
      "loss": 0.1041,
      "num_input_tokens_seen": 61092824,
      "step": 37975
    },
    {
      "epoch": 1.7193462957513752,
      "grad_norm": 0.8234824538230896,
      "learning_rate": 3.1427539284736297e-07,
      "loss": 0.1458,
      "num_input_tokens_seen": 61100472,
      "step": 37980
    },
    {
      "epoch": 1.7195726476380182,
      "grad_norm": 0.3193970322608948,
      "learning_rate": 3.127255136733093e-07,
      "loss": 0.1086,
      "num_input_tokens_seen": 61108760,
      "step": 37985
    },
    {
      "epoch": 1.7197989995246612,
      "grad_norm": 0.531320333480835,
      "learning_rate": 3.1117944158722544e-07,
      "loss": 0.1571,
      "num_input_tokens_seen": 61116600,
      "step": 37990
    },
    {
      "epoch": 1.720025351411304,
      "grad_norm": 1.410086989402771,
      "learning_rate": 3.0963717682752635e-07,
      "loss": 0.1577,
      "num_input_tokens_seen": 61124504,
      "step": 37995
    },
    {
      "epoch": 1.720251703297947,
      "grad_norm": 0.3368818759918213,
      "learning_rate": 3.080987196320578e-07,
      "loss": 0.1189,
      "num_input_tokens_seen": 61132280,
      "step": 38000
    },
    {
      "epoch": 1.720251703297947,
      "eval_loss": 0.14362019300460815,
      "eval_runtime": 405.3497,
      "eval_samples_per_second": 96.882,
      "eval_steps_per_second": 24.221,
      "num_input_tokens_seen": 61132280,
      "step": 38000
    },
    {
      "epoch": 1.72047805518459,
      "grad_norm": 0.7167856693267822,
      "learning_rate": 3.065640702380607e-07,
      "loss": 0.1372,
      "num_input_tokens_seen": 61140280,
      "step": 38005
    },
    {
      "epoch": 1.720704407071233,
      "grad_norm": 0.5273398756980896,
      "learning_rate": 3.050332288822011e-07,
      "loss": 0.1152,
      "num_input_tokens_seen": 61147992,
      "step": 38010
    },
    {
      "epoch": 1.720930758957876,
      "grad_norm": 1.2360140085220337,
      "learning_rate": 3.035061958005542e-07,
      "loss": 0.158,
      "num_input_tokens_seen": 61157080,
      "step": 38015
    },
    {
      "epoch": 1.7211571108445187,
      "grad_norm": 0.6596002578735352,
      "learning_rate": 3.019829712286093e-07,
      "loss": 0.1321,
      "num_input_tokens_seen": 61165496,
      "step": 38020
    },
    {
      "epoch": 1.7213834627311617,
      "grad_norm": 0.3969244956970215,
      "learning_rate": 3.004635554012647e-07,
      "loss": 0.118,
      "num_input_tokens_seen": 61173688,
      "step": 38025
    },
    {
      "epoch": 1.7216098146178047,
      "grad_norm": 0.7611373662948608,
      "learning_rate": 2.9894794855283017e-07,
      "loss": 0.1461,
      "num_input_tokens_seen": 61181624,
      "step": 38030
    },
    {
      "epoch": 1.7218361665044477,
      "grad_norm": 0.6667451858520508,
      "learning_rate": 2.9743615091703816e-07,
      "loss": 0.1212,
      "num_input_tokens_seen": 61189656,
      "step": 38035
    },
    {
      "epoch": 1.7220625183910907,
      "grad_norm": 0.8209425806999207,
      "learning_rate": 2.959281627270216e-07,
      "loss": 0.1153,
      "num_input_tokens_seen": 61197368,
      "step": 38040
    },
    {
      "epoch": 1.7222888702777337,
      "grad_norm": 0.3893747925758362,
      "learning_rate": 2.944239842153362e-07,
      "loss": 0.1249,
      "num_input_tokens_seen": 61205528,
      "step": 38045
    },
    {
      "epoch": 1.7225152221643767,
      "grad_norm": 0.5193571448326111,
      "learning_rate": 2.929236156139381e-07,
      "loss": 0.125,
      "num_input_tokens_seen": 61213880,
      "step": 38050
    },
    {
      "epoch": 1.7227415740510197,
      "grad_norm": 0.4197123050689697,
      "learning_rate": 2.9142705715420883e-07,
      "loss": 0.1568,
      "num_input_tokens_seen": 61222072,
      "step": 38055
    },
    {
      "epoch": 1.7229679259376627,
      "grad_norm": 0.9361463189125061,
      "learning_rate": 2.8993430906693595e-07,
      "loss": 0.1809,
      "num_input_tokens_seen": 61230232,
      "step": 38060
    },
    {
      "epoch": 1.7231942778243057,
      "grad_norm": 0.7898455858230591,
      "learning_rate": 2.88445371582316e-07,
      "loss": 0.1306,
      "num_input_tokens_seen": 61238168,
      "step": 38065
    },
    {
      "epoch": 1.7234206297109487,
      "grad_norm": 0.40540945529937744,
      "learning_rate": 2.8696024492996796e-07,
      "loss": 0.1117,
      "num_input_tokens_seen": 61246264,
      "step": 38070
    },
    {
      "epoch": 1.7236469815975917,
      "grad_norm": 0.9152389764785767,
      "learning_rate": 2.854789293389115e-07,
      "loss": 0.1236,
      "num_input_tokens_seen": 61254456,
      "step": 38075
    },
    {
      "epoch": 1.7238733334842347,
      "grad_norm": 1.2826964855194092,
      "learning_rate": 2.8400142503758606e-07,
      "loss": 0.1853,
      "num_input_tokens_seen": 61262872,
      "step": 38080
    },
    {
      "epoch": 1.7240996853708777,
      "grad_norm": 1.215058445930481,
      "learning_rate": 2.8252773225384276e-07,
      "loss": 0.1849,
      "num_input_tokens_seen": 61270904,
      "step": 38085
    },
    {
      "epoch": 1.7243260372575206,
      "grad_norm": 0.7824463248252869,
      "learning_rate": 2.8105785121494143e-07,
      "loss": 0.122,
      "num_input_tokens_seen": 61278456,
      "step": 38090
    },
    {
      "epoch": 1.7245523891441636,
      "grad_norm": 1.0415230989456177,
      "learning_rate": 2.795917821475563e-07,
      "loss": 0.143,
      "num_input_tokens_seen": 61286456,
      "step": 38095
    },
    {
      "epoch": 1.7247787410308066,
      "grad_norm": 0.547629177570343,
      "learning_rate": 2.78129525277776e-07,
      "loss": 0.1374,
      "num_input_tokens_seen": 61295064,
      "step": 38100
    },
    {
      "epoch": 1.7250050929174494,
      "grad_norm": 0.18302124738693237,
      "learning_rate": 2.766710808310952e-07,
      "loss": 0.1049,
      "num_input_tokens_seen": 61303320,
      "step": 38105
    },
    {
      "epoch": 1.7252314448040924,
      "grad_norm": 1.3470631837844849,
      "learning_rate": 2.7521644903242827e-07,
      "loss": 0.1576,
      "num_input_tokens_seen": 61311576,
      "step": 38110
    },
    {
      "epoch": 1.7254577966907354,
      "grad_norm": 0.44496312737464905,
      "learning_rate": 2.7376563010609593e-07,
      "loss": 0.152,
      "num_input_tokens_seen": 61319448,
      "step": 38115
    },
    {
      "epoch": 1.7256841485773784,
      "grad_norm": 0.8640407919883728,
      "learning_rate": 2.72318624275833e-07,
      "loss": 0.1322,
      "num_input_tokens_seen": 61327064,
      "step": 38120
    },
    {
      "epoch": 1.7259105004640214,
      "grad_norm": 0.7075752019882202,
      "learning_rate": 2.7087543176478324e-07,
      "loss": 0.1468,
      "num_input_tokens_seen": 61334968,
      "step": 38125
    },
    {
      "epoch": 1.7261368523506644,
      "grad_norm": 0.5197558999061584,
      "learning_rate": 2.694360527955103e-07,
      "loss": 0.1716,
      "num_input_tokens_seen": 61342968,
      "step": 38130
    },
    {
      "epoch": 1.7263632042373072,
      "grad_norm": 1.036697506904602,
      "learning_rate": 2.680004875899811e-07,
      "loss": 0.1256,
      "num_input_tokens_seen": 61351416,
      "step": 38135
    },
    {
      "epoch": 1.7265895561239502,
      "grad_norm": 0.3622368276119232,
      "learning_rate": 2.665687363695768e-07,
      "loss": 0.1236,
      "num_input_tokens_seen": 61359224,
      "step": 38140
    },
    {
      "epoch": 1.7268159080105931,
      "grad_norm": 0.8804404139518738,
      "learning_rate": 2.6514079935509584e-07,
      "loss": 0.1901,
      "num_input_tokens_seen": 61367864,
      "step": 38145
    },
    {
      "epoch": 1.7270422598972361,
      "grad_norm": 0.7519844174385071,
      "learning_rate": 2.6371667676673983e-07,
      "loss": 0.1451,
      "num_input_tokens_seen": 61375832,
      "step": 38150
    },
    {
      "epoch": 1.7272686117838791,
      "grad_norm": 0.4785972535610199,
      "learning_rate": 2.6229636882412755e-07,
      "loss": 0.1461,
      "num_input_tokens_seen": 61383448,
      "step": 38155
    },
    {
      "epoch": 1.7274949636705221,
      "grad_norm": 0.27251511812210083,
      "learning_rate": 2.6087987574628935e-07,
      "loss": 0.1101,
      "num_input_tokens_seen": 61391832,
      "step": 38160
    },
    {
      "epoch": 1.7277213155571651,
      "grad_norm": 0.7661013007164001,
      "learning_rate": 2.5946719775166437e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 61400344,
      "step": 38165
    },
    {
      "epoch": 1.7279476674438081,
      "grad_norm": 0.4370274245738983,
      "learning_rate": 2.5805833505810616e-07,
      "loss": 0.1505,
      "num_input_tokens_seen": 61408440,
      "step": 38170
    },
    {
      "epoch": 1.7281740193304511,
      "grad_norm": 0.5007676482200623,
      "learning_rate": 2.566532878828798e-07,
      "loss": 0.1665,
      "num_input_tokens_seen": 61417016,
      "step": 38175
    },
    {
      "epoch": 1.7284003712170941,
      "grad_norm": 0.7088289260864258,
      "learning_rate": 2.552520564426619e-07,
      "loss": 0.1641,
      "num_input_tokens_seen": 61424760,
      "step": 38180
    },
    {
      "epoch": 1.728626723103737,
      "grad_norm": 0.6435517072677612,
      "learning_rate": 2.5385464095353803e-07,
      "loss": 0.1848,
      "num_input_tokens_seen": 61432600,
      "step": 38185
    },
    {
      "epoch": 1.72885307499038,
      "grad_norm": 0.5882324576377869,
      "learning_rate": 2.5246104163100804e-07,
      "loss": 0.1286,
      "num_input_tokens_seen": 61440792,
      "step": 38190
    },
    {
      "epoch": 1.729079426877023,
      "grad_norm": 1.2219350337982178,
      "learning_rate": 2.510712586899833e-07,
      "loss": 0.1459,
      "num_input_tokens_seen": 61448312,
      "step": 38195
    },
    {
      "epoch": 1.729305778763666,
      "grad_norm": 0.7735883593559265,
      "learning_rate": 2.4968529234478124e-07,
      "loss": 0.1546,
      "num_input_tokens_seen": 61456312,
      "step": 38200
    },
    {
      "epoch": 1.729305778763666,
      "eval_loss": 0.14321373403072357,
      "eval_runtime": 404.7191,
      "eval_samples_per_second": 97.033,
      "eval_steps_per_second": 24.259,
      "num_input_tokens_seen": 61456312,
      "step": 38200
    },
    {
      "epoch": 1.729532130650309,
      "grad_norm": 0.6687813997268677,
      "learning_rate": 2.483031428091448e-07,
      "loss": 0.1315,
      "num_input_tokens_seen": 61464024,
      "step": 38205
    },
    {
      "epoch": 1.729758482536952,
      "grad_norm": 0.9812148809432983,
      "learning_rate": 2.469248102962091e-07,
      "loss": 0.1088,
      "num_input_tokens_seen": 61471448,
      "step": 38210
    },
    {
      "epoch": 1.729984834423595,
      "grad_norm": 0.572920024394989,
      "learning_rate": 2.4555029501853455e-07,
      "loss": 0.1628,
      "num_input_tokens_seen": 61479384,
      "step": 38215
    },
    {
      "epoch": 1.7302111863102378,
      "grad_norm": 0.4037631154060364,
      "learning_rate": 2.441795971880906e-07,
      "loss": 0.1476,
      "num_input_tokens_seen": 61487736,
      "step": 38220
    },
    {
      "epoch": 1.7304375381968808,
      "grad_norm": 1.4874355792999268,
      "learning_rate": 2.4281271701625255e-07,
      "loss": 0.1229,
      "num_input_tokens_seen": 61496344,
      "step": 38225
    },
    {
      "epoch": 1.7306638900835238,
      "grad_norm": 0.6732045412063599,
      "learning_rate": 2.4144965471381007e-07,
      "loss": 0.123,
      "num_input_tokens_seen": 61504280,
      "step": 38230
    },
    {
      "epoch": 1.7308902419701668,
      "grad_norm": 0.5955372452735901,
      "learning_rate": 2.400904104909674e-07,
      "loss": 0.1379,
      "num_input_tokens_seen": 61513048,
      "step": 38235
    },
    {
      "epoch": 1.7311165938568098,
      "grad_norm": 0.8533313274383545,
      "learning_rate": 2.3873498455733725e-07,
      "loss": 0.1731,
      "num_input_tokens_seen": 61521208,
      "step": 38240
    },
    {
      "epoch": 1.7313429457434528,
      "grad_norm": 0.3140442967414856,
      "learning_rate": 2.3738337712194137e-07,
      "loss": 0.142,
      "num_input_tokens_seen": 61529016,
      "step": 38245
    },
    {
      "epoch": 1.7315692976300956,
      "grad_norm": 0.6006019115447998,
      "learning_rate": 2.3603558839321305e-07,
      "loss": 0.1233,
      "num_input_tokens_seen": 61537592,
      "step": 38250
    },
    {
      "epoch": 1.7317956495167386,
      "grad_norm": 0.5917344689369202,
      "learning_rate": 2.3469161857900267e-07,
      "loss": 0.1437,
      "num_input_tokens_seen": 61545112,
      "step": 38255
    },
    {
      "epoch": 1.7320220014033816,
      "grad_norm": 0.5078831315040588,
      "learning_rate": 2.3335146788656393e-07,
      "loss": 0.1304,
      "num_input_tokens_seen": 61552504,
      "step": 38260
    },
    {
      "epoch": 1.7322483532900246,
      "grad_norm": 0.8974546790122986,
      "learning_rate": 2.3201513652256757e-07,
      "loss": 0.1632,
      "num_input_tokens_seen": 61560824,
      "step": 38265
    },
    {
      "epoch": 1.7324747051766676,
      "grad_norm": 0.24678368866443634,
      "learning_rate": 2.3068262469308766e-07,
      "loss": 0.0997,
      "num_input_tokens_seen": 61569080,
      "step": 38270
    },
    {
      "epoch": 1.7327010570633106,
      "grad_norm": 0.6190410256385803,
      "learning_rate": 2.2935393260362093e-07,
      "loss": 0.155,
      "num_input_tokens_seen": 61577144,
      "step": 38275
    },
    {
      "epoch": 1.7329274089499536,
      "grad_norm": 0.9091770052909851,
      "learning_rate": 2.2802906045906458e-07,
      "loss": 0.1123,
      "num_input_tokens_seen": 61585112,
      "step": 38280
    },
    {
      "epoch": 1.7331537608365966,
      "grad_norm": 0.9104623794555664,
      "learning_rate": 2.2670800846373018e-07,
      "loss": 0.1506,
      "num_input_tokens_seen": 61593304,
      "step": 38285
    },
    {
      "epoch": 1.7333801127232396,
      "grad_norm": 0.9603397846221924,
      "learning_rate": 2.2539077682134367e-07,
      "loss": 0.1664,
      "num_input_tokens_seen": 61601400,
      "step": 38290
    },
    {
      "epoch": 1.7336064646098825,
      "grad_norm": 0.7990619540214539,
      "learning_rate": 2.2407736573503423e-07,
      "loss": 0.1289,
      "num_input_tokens_seen": 61609528,
      "step": 38295
    },
    {
      "epoch": 1.7338328164965255,
      "grad_norm": 0.8086264133453369,
      "learning_rate": 2.2276777540735093e-07,
      "loss": 0.164,
      "num_input_tokens_seen": 61617912,
      "step": 38300
    },
    {
      "epoch": 1.7340591683831685,
      "grad_norm": 0.5574609637260437,
      "learning_rate": 2.2146200604024613e-07,
      "loss": 0.1137,
      "num_input_tokens_seen": 61626328,
      "step": 38305
    },
    {
      "epoch": 1.7342855202698115,
      "grad_norm": 0.928867757320404,
      "learning_rate": 2.2016005783508375e-07,
      "loss": 0.1531,
      "num_input_tokens_seen": 61634296,
      "step": 38310
    },
    {
      "epoch": 1.7345118721564545,
      "grad_norm": 0.5154803991317749,
      "learning_rate": 2.1886193099264763e-07,
      "loss": 0.1345,
      "num_input_tokens_seen": 61642872,
      "step": 38315
    },
    {
      "epoch": 1.7347382240430975,
      "grad_norm": 0.5846201181411743,
      "learning_rate": 2.175676257131165e-07,
      "loss": 0.1283,
      "num_input_tokens_seen": 61650648,
      "step": 38320
    },
    {
      "epoch": 1.7349645759297405,
      "grad_norm": 1.3683452606201172,
      "learning_rate": 2.162771421960974e-07,
      "loss": 0.1721,
      "num_input_tokens_seen": 61658488,
      "step": 38325
    },
    {
      "epoch": 1.7351909278163835,
      "grad_norm": 0.22213956713676453,
      "learning_rate": 2.1499048064059224e-07,
      "loss": 0.1292,
      "num_input_tokens_seen": 61666776,
      "step": 38330
    },
    {
      "epoch": 1.7354172797030263,
      "grad_norm": 0.38587021827697754,
      "learning_rate": 2.1370764124502285e-07,
      "loss": 0.1036,
      "num_input_tokens_seen": 61674296,
      "step": 38335
    },
    {
      "epoch": 1.7356436315896693,
      "grad_norm": 0.4464815855026245,
      "learning_rate": 2.1242862420721988e-07,
      "loss": 0.1116,
      "num_input_tokens_seen": 61681848,
      "step": 38340
    },
    {
      "epoch": 1.7358699834763123,
      "grad_norm": 0.8522152900695801,
      "learning_rate": 2.1115342972442276e-07,
      "loss": 0.1354,
      "num_input_tokens_seen": 61689752,
      "step": 38345
    },
    {
      "epoch": 1.7360963353629553,
      "grad_norm": 0.6935303211212158,
      "learning_rate": 2.0988205799328252e-07,
      "loss": 0.1461,
      "num_input_tokens_seen": 61697848,
      "step": 38350
    },
    {
      "epoch": 1.7363226872495983,
      "grad_norm": 0.6839206218719482,
      "learning_rate": 2.0861450920986182e-07,
      "loss": 0.1905,
      "num_input_tokens_seen": 61706232,
      "step": 38355
    },
    {
      "epoch": 1.736549039136241,
      "grad_norm": 0.9022559523582458,
      "learning_rate": 2.07350783569632e-07,
      "loss": 0.1452,
      "num_input_tokens_seen": 61713848,
      "step": 38360
    },
    {
      "epoch": 1.736775391022884,
      "grad_norm": 0.29102060198783875,
      "learning_rate": 2.060908812674761e-07,
      "loss": 0.1329,
      "num_input_tokens_seen": 61721880,
      "step": 38365
    },
    {
      "epoch": 1.737001742909527,
      "grad_norm": 0.6591131091117859,
      "learning_rate": 2.0483480249768317e-07,
      "loss": 0.149,
      "num_input_tokens_seen": 61730136,
      "step": 38370
    },
    {
      "epoch": 1.73722809479617,
      "grad_norm": 0.7321729063987732,
      "learning_rate": 2.035825474539621e-07,
      "loss": 0.1704,
      "num_input_tokens_seen": 61737752,
      "step": 38375
    },
    {
      "epoch": 1.737454446682813,
      "grad_norm": 0.7387709617614746,
      "learning_rate": 2.0233411632942235e-07,
      "loss": 0.1652,
      "num_input_tokens_seen": 61745592,
      "step": 38380
    },
    {
      "epoch": 1.737680798569456,
      "grad_norm": 0.719001054763794,
      "learning_rate": 2.0108950931658764e-07,
      "loss": 0.1326,
      "num_input_tokens_seen": 61753592,
      "step": 38385
    },
    {
      "epoch": 1.737907150456099,
      "grad_norm": 0.8033308386802673,
      "learning_rate": 1.998487266073934e-07,
      "loss": 0.207,
      "num_input_tokens_seen": 61761368,
      "step": 38390
    },
    {
      "epoch": 1.738133502342742,
      "grad_norm": 0.6230449676513672,
      "learning_rate": 1.986117683931865e-07,
      "loss": 0.1946,
      "num_input_tokens_seen": 61769208,
      "step": 38395
    },
    {
      "epoch": 1.738359854229385,
      "grad_norm": 1.251970648765564,
      "learning_rate": 1.9737863486471442e-07,
      "loss": 0.1326,
      "num_input_tokens_seen": 61776888,
      "step": 38400
    },
    {
      "epoch": 1.738359854229385,
      "eval_loss": 0.14356054365634918,
      "eval_runtime": 404.6597,
      "eval_samples_per_second": 97.047,
      "eval_steps_per_second": 24.262,
      "num_input_tokens_seen": 61776888,
      "step": 38400
    },
    {
      "epoch": 1.738586206116028,
      "grad_norm": 0.5242164134979248,
      "learning_rate": 1.9614932621215e-07,
      "loss": 0.1896,
      "num_input_tokens_seen": 61785560,
      "step": 38405
    },
    {
      "epoch": 1.738812558002671,
      "grad_norm": 0.6411065459251404,
      "learning_rate": 1.9492384262506102e-07,
      "loss": 0.1301,
      "num_input_tokens_seen": 61793176,
      "step": 38410
    },
    {
      "epoch": 1.739038909889314,
      "grad_norm": 0.35677510499954224,
      "learning_rate": 1.9370218429243524e-07,
      "loss": 0.1326,
      "num_input_tokens_seen": 61801400,
      "step": 38415
    },
    {
      "epoch": 1.739265261775957,
      "grad_norm": 0.4917336702346802,
      "learning_rate": 1.9248435140267197e-07,
      "loss": 0.1316,
      "num_input_tokens_seen": 61808984,
      "step": 38420
    },
    {
      "epoch": 1.7394916136626,
      "grad_norm": 0.6793871521949768,
      "learning_rate": 1.9127034414356814e-07,
      "loss": 0.1318,
      "num_input_tokens_seen": 61816856,
      "step": 38425
    },
    {
      "epoch": 1.739717965549243,
      "grad_norm": 1.0591514110565186,
      "learning_rate": 1.9006016270234627e-07,
      "loss": 0.1372,
      "num_input_tokens_seen": 61824632,
      "step": 38430
    },
    {
      "epoch": 1.739944317435886,
      "grad_norm": 0.4731978178024292,
      "learning_rate": 1.888538072656293e-07,
      "loss": 0.1228,
      "num_input_tokens_seen": 61833112,
      "step": 38435
    },
    {
      "epoch": 1.740170669322529,
      "grad_norm": 0.4239657521247864,
      "learning_rate": 1.8765127801944893e-07,
      "loss": 0.1029,
      "num_input_tokens_seen": 61840728,
      "step": 38440
    },
    {
      "epoch": 1.7403970212091717,
      "grad_norm": 0.2952893376350403,
      "learning_rate": 1.8645257514925406e-07,
      "loss": 0.1294,
      "num_input_tokens_seen": 61849176,
      "step": 38445
    },
    {
      "epoch": 1.7406233730958147,
      "grad_norm": 0.8844936490058899,
      "learning_rate": 1.8525769883989685e-07,
      "loss": 0.1443,
      "num_input_tokens_seen": 61856376,
      "step": 38450
    },
    {
      "epoch": 1.7408497249824577,
      "grad_norm": 1.0146517753601074,
      "learning_rate": 1.8406664927564654e-07,
      "loss": 0.1616,
      "num_input_tokens_seen": 61863288,
      "step": 38455
    },
    {
      "epoch": 1.7410760768691007,
      "grad_norm": 1.4316041469573975,
      "learning_rate": 1.8287942664017566e-07,
      "loss": 0.1514,
      "num_input_tokens_seen": 61871160,
      "step": 38460
    },
    {
      "epoch": 1.7413024287557437,
      "grad_norm": 0.7608212232589722,
      "learning_rate": 1.8169603111656552e-07,
      "loss": 0.1519,
      "num_input_tokens_seen": 61879032,
      "step": 38465
    },
    {
      "epoch": 1.7415287806423867,
      "grad_norm": 0.3519236743450165,
      "learning_rate": 1.805164628873146e-07,
      "loss": 0.12,
      "num_input_tokens_seen": 61886968,
      "step": 38470
    },
    {
      "epoch": 1.7417551325290295,
      "grad_norm": 0.31282538175582886,
      "learning_rate": 1.793407221343274e-07,
      "loss": 0.1404,
      "num_input_tokens_seen": 61894744,
      "step": 38475
    },
    {
      "epoch": 1.7419814844156725,
      "grad_norm": 0.6178083419799805,
      "learning_rate": 1.781688090389172e-07,
      "loss": 0.1193,
      "num_input_tokens_seen": 61903352,
      "step": 38480
    },
    {
      "epoch": 1.7422078363023155,
      "grad_norm": 1.0206537246704102,
      "learning_rate": 1.770007237818061e-07,
      "loss": 0.1536,
      "num_input_tokens_seen": 61911256,
      "step": 38485
    },
    {
      "epoch": 1.7424341881889585,
      "grad_norm": 0.9649876356124878,
      "learning_rate": 1.7583646654313059e-07,
      "loss": 0.109,
      "num_input_tokens_seen": 61918840,
      "step": 38490
    },
    {
      "epoch": 1.7426605400756014,
      "grad_norm": 0.995634138584137,
      "learning_rate": 1.7467603750242757e-07,
      "loss": 0.1095,
      "num_input_tokens_seen": 61926200,
      "step": 38495
    },
    {
      "epoch": 1.7428868919622444,
      "grad_norm": 1.0825705528259277,
      "learning_rate": 1.7351943683865944e-07,
      "loss": 0.1586,
      "num_input_tokens_seen": 61934328,
      "step": 38500
    },
    {
      "epoch": 1.7431132438488874,
      "grad_norm": 0.5240523219108582,
      "learning_rate": 1.723666647301808e-07,
      "loss": 0.1529,
      "num_input_tokens_seen": 61942360,
      "step": 38505
    },
    {
      "epoch": 1.7433395957355304,
      "grad_norm": 0.7480849027633667,
      "learning_rate": 1.712177213547661e-07,
      "loss": 0.1378,
      "num_input_tokens_seen": 61949944,
      "step": 38510
    },
    {
      "epoch": 1.7435659476221734,
      "grad_norm": 0.7633200287818909,
      "learning_rate": 1.7007260688959581e-07,
      "loss": 0.1425,
      "num_input_tokens_seen": 61958424,
      "step": 38515
    },
    {
      "epoch": 1.7437922995088164,
      "grad_norm": 0.3076753616333008,
      "learning_rate": 1.68931321511262e-07,
      "loss": 0.1465,
      "num_input_tokens_seen": 61966712,
      "step": 38520
    },
    {
      "epoch": 1.7440186513954594,
      "grad_norm": 0.48506173491477966,
      "learning_rate": 1.6779386539576835e-07,
      "loss": 0.127,
      "num_input_tokens_seen": 61974680,
      "step": 38525
    },
    {
      "epoch": 1.7442450032821024,
      "grad_norm": 0.7736327648162842,
      "learning_rate": 1.666602387185162e-07,
      "loss": 0.1665,
      "num_input_tokens_seen": 61982584,
      "step": 38530
    },
    {
      "epoch": 1.7444713551687454,
      "grad_norm": 0.2768321931362152,
      "learning_rate": 1.655304416543352e-07,
      "loss": 0.1153,
      "num_input_tokens_seen": 61990680,
      "step": 38535
    },
    {
      "epoch": 1.7446977070553884,
      "grad_norm": 1.2897493839263916,
      "learning_rate": 1.6440447437744698e-07,
      "loss": 0.1164,
      "num_input_tokens_seen": 61998392,
      "step": 38540
    },
    {
      "epoch": 1.7449240589420314,
      "grad_norm": 0.4977518320083618,
      "learning_rate": 1.6328233706149332e-07,
      "loss": 0.1407,
      "num_input_tokens_seen": 62006424,
      "step": 38545
    },
    {
      "epoch": 1.7451504108286744,
      "grad_norm": 0.8212196826934814,
      "learning_rate": 1.6216402987951906e-07,
      "loss": 0.1289,
      "num_input_tokens_seen": 62014296,
      "step": 38550
    },
    {
      "epoch": 1.7453767627153174,
      "grad_norm": 1.7077876329421997,
      "learning_rate": 1.6104955300398627e-07,
      "loss": 0.1031,
      "num_input_tokens_seen": 62022264,
      "step": 38555
    },
    {
      "epoch": 1.7456031146019602,
      "grad_norm": 1.0565440654754639,
      "learning_rate": 1.5993890660675748e-07,
      "loss": 0.1655,
      "num_input_tokens_seen": 62030072,
      "step": 38560
    },
    {
      "epoch": 1.7458294664886032,
      "grad_norm": 0.4485406279563904,
      "learning_rate": 1.5883209085910678e-07,
      "loss": 0.1489,
      "num_input_tokens_seen": 62038424,
      "step": 38565
    },
    {
      "epoch": 1.7460558183752461,
      "grad_norm": 0.5925173759460449,
      "learning_rate": 1.5772910593172264e-07,
      "loss": 0.135,
      "num_input_tokens_seen": 62046520,
      "step": 38570
    },
    {
      "epoch": 1.7462821702618891,
      "grad_norm": 0.37189823389053345,
      "learning_rate": 1.5662995199469954e-07,
      "loss": 0.1759,
      "num_input_tokens_seen": 62054104,
      "step": 38575
    },
    {
      "epoch": 1.7465085221485321,
      "grad_norm": 0.34047913551330566,
      "learning_rate": 1.5553462921753802e-07,
      "loss": 0.0866,
      "num_input_tokens_seen": 62062168,
      "step": 38580
    },
    {
      "epoch": 1.7467348740351751,
      "grad_norm": 0.7735950946807861,
      "learning_rate": 1.544431377691502e-07,
      "loss": 0.1446,
      "num_input_tokens_seen": 62070264,
      "step": 38585
    },
    {
      "epoch": 1.746961225921818,
      "grad_norm": 0.8205127120018005,
      "learning_rate": 1.5335547781785975e-07,
      "loss": 0.1384,
      "num_input_tokens_seen": 62077784,
      "step": 38590
    },
    {
      "epoch": 1.747187577808461,
      "grad_norm": 0.6823034882545471,
      "learning_rate": 1.5227164953139917e-07,
      "loss": 0.1192,
      "num_input_tokens_seen": 62085464,
      "step": 38595
    },
    {
      "epoch": 1.747413929695104,
      "grad_norm": 0.7783083319664001,
      "learning_rate": 1.511916530769042e-07,
      "loss": 0.1584,
      "num_input_tokens_seen": 62093144,
      "step": 38600
    },
    {
      "epoch": 1.747413929695104,
      "eval_loss": 0.14358901977539062,
      "eval_runtime": 405.9644,
      "eval_samples_per_second": 96.735,
      "eval_steps_per_second": 24.184,
      "num_input_tokens_seen": 62093144,
      "step": 38600
    },
    {
      "epoch": 1.747640281581747,
      "grad_norm": 0.5343934297561646,
      "learning_rate": 1.5011548862092773e-07,
      "loss": 0.1689,
      "num_input_tokens_seen": 62100728,
      "step": 38605
    },
    {
      "epoch": 1.7478666334683899,
      "grad_norm": 0.29310157895088196,
      "learning_rate": 1.490431563294231e-07,
      "loss": 0.1146,
      "num_input_tokens_seen": 62108024,
      "step": 38610
    },
    {
      "epoch": 1.7480929853550329,
      "grad_norm": 0.9500014781951904,
      "learning_rate": 1.4797465636776365e-07,
      "loss": 0.1173,
      "num_input_tokens_seen": 62115928,
      "step": 38615
    },
    {
      "epoch": 1.7483193372416759,
      "grad_norm": 0.8233439326286316,
      "learning_rate": 1.4690998890072027e-07,
      "loss": 0.1592,
      "num_input_tokens_seen": 62123864,
      "step": 38620
    },
    {
      "epoch": 1.7485456891283189,
      "grad_norm": 0.4164626896381378,
      "learning_rate": 1.4584915409248112e-07,
      "loss": 0.1363,
      "num_input_tokens_seen": 62131704,
      "step": 38625
    },
    {
      "epoch": 1.7487720410149619,
      "grad_norm": 0.48793578147888184,
      "learning_rate": 1.4479215210663754e-07,
      "loss": 0.1354,
      "num_input_tokens_seen": 62138936,
      "step": 38630
    },
    {
      "epoch": 1.7489983929016049,
      "grad_norm": 0.6912071108818054,
      "learning_rate": 1.4373898310619528e-07,
      "loss": 0.1431,
      "num_input_tokens_seen": 62146808,
      "step": 38635
    },
    {
      "epoch": 1.7492247447882479,
      "grad_norm": 1.195762276649475,
      "learning_rate": 1.4268964725356604e-07,
      "loss": 0.1656,
      "num_input_tokens_seen": 62155672,
      "step": 38640
    },
    {
      "epoch": 1.7494510966748908,
      "grad_norm": 0.638842761516571,
      "learning_rate": 1.4164414471056764e-07,
      "loss": 0.1517,
      "num_input_tokens_seen": 62163864,
      "step": 38645
    },
    {
      "epoch": 1.7496774485615338,
      "grad_norm": 0.3540121912956238,
      "learning_rate": 1.4060247563843497e-07,
      "loss": 0.1319,
      "num_input_tokens_seen": 62171960,
      "step": 38650
    },
    {
      "epoch": 1.7499038004481768,
      "grad_norm": 0.6312305927276611,
      "learning_rate": 1.3956464019780068e-07,
      "loss": 0.1278,
      "num_input_tokens_seen": 62179864,
      "step": 38655
    },
    {
      "epoch": 1.7501301523348198,
      "grad_norm": 0.6370823383331299,
      "learning_rate": 1.385306385487145e-07,
      "loss": 0.1446,
      "num_input_tokens_seen": 62188088,
      "step": 38660
    },
    {
      "epoch": 1.7503565042214628,
      "grad_norm": 0.4276919662952423,
      "learning_rate": 1.3750047085063222e-07,
      "loss": 0.1,
      "num_input_tokens_seen": 62195384,
      "step": 38665
    },
    {
      "epoch": 1.7505828561081058,
      "grad_norm": 0.6607746481895447,
      "learning_rate": 1.3647413726242119e-07,
      "loss": 0.1711,
      "num_input_tokens_seen": 62203096,
      "step": 38670
    },
    {
      "epoch": 1.7508092079947486,
      "grad_norm": 0.36910340189933777,
      "learning_rate": 1.3545163794235205e-07,
      "loss": 0.1348,
      "num_input_tokens_seen": 62211160,
      "step": 38675
    },
    {
      "epoch": 1.7510355598813916,
      "grad_norm": 0.6998895406723022,
      "learning_rate": 1.3443297304810698e-07,
      "loss": 0.1283,
      "num_input_tokens_seen": 62218584,
      "step": 38680
    },
    {
      "epoch": 1.7512619117680346,
      "grad_norm": 0.69915372133255,
      "learning_rate": 1.3341814273677977e-07,
      "loss": 0.1822,
      "num_input_tokens_seen": 62226584,
      "step": 38685
    },
    {
      "epoch": 1.7514882636546776,
      "grad_norm": 0.5925779938697815,
      "learning_rate": 1.324071471648647e-07,
      "loss": 0.1179,
      "num_input_tokens_seen": 62235288,
      "step": 38690
    },
    {
      "epoch": 1.7517146155413206,
      "grad_norm": 0.45576369762420654,
      "learning_rate": 1.3139998648827312e-07,
      "loss": 0.1269,
      "num_input_tokens_seen": 62242744,
      "step": 38695
    },
    {
      "epoch": 1.7519409674279633,
      "grad_norm": 0.6054589748382568,
      "learning_rate": 1.3039666086232526e-07,
      "loss": 0.1268,
      "num_input_tokens_seen": 62250520,
      "step": 38700
    },
    {
      "epoch": 1.7521673193146063,
      "grad_norm": 1.0169627666473389,
      "learning_rate": 1.2939717044174183e-07,
      "loss": 0.1501,
      "num_input_tokens_seen": 62258456,
      "step": 38705
    },
    {
      "epoch": 1.7523936712012493,
      "grad_norm": 0.6999267935752869,
      "learning_rate": 1.284015153806578e-07,
      "loss": 0.1575,
      "num_input_tokens_seen": 62266104,
      "step": 38710
    },
    {
      "epoch": 1.7526200230878923,
      "grad_norm": 0.5312695503234863,
      "learning_rate": 1.274096958326171e-07,
      "loss": 0.1194,
      "num_input_tokens_seen": 62273944,
      "step": 38715
    },
    {
      "epoch": 1.7528463749745353,
      "grad_norm": 0.8057281970977783,
      "learning_rate": 1.2642171195056952e-07,
      "loss": 0.1436,
      "num_input_tokens_seen": 62281816,
      "step": 38720
    },
    {
      "epoch": 1.7530727268611783,
      "grad_norm": 0.8651970028877258,
      "learning_rate": 1.2543756388687377e-07,
      "loss": 0.149,
      "num_input_tokens_seen": 62290328,
      "step": 38725
    },
    {
      "epoch": 1.7532990787478213,
      "grad_norm": 0.7375943064689636,
      "learning_rate": 1.2445725179330014e-07,
      "loss": 0.1636,
      "num_input_tokens_seen": 62298328,
      "step": 38730
    },
    {
      "epoch": 1.7535254306344643,
      "grad_norm": 0.6621701717376709,
      "learning_rate": 1.2348077582102212e-07,
      "loss": 0.1247,
      "num_input_tokens_seen": 62306136,
      "step": 38735
    },
    {
      "epoch": 1.7537517825211073,
      "grad_norm": 0.9286881685256958,
      "learning_rate": 1.2250813612062762e-07,
      "loss": 0.138,
      "num_input_tokens_seen": 62314136,
      "step": 38740
    },
    {
      "epoch": 1.7539781344077503,
      "grad_norm": 0.5867607593536377,
      "learning_rate": 1.215393328421105e-07,
      "loss": 0.1821,
      "num_input_tokens_seen": 62322136,
      "step": 38745
    },
    {
      "epoch": 1.7542044862943933,
      "grad_norm": 0.5627220869064331,
      "learning_rate": 1.2057436613486796e-07,
      "loss": 0.169,
      "num_input_tokens_seen": 62329912,
      "step": 38750
    },
    {
      "epoch": 1.7544308381810363,
      "grad_norm": 0.9580448269844055,
      "learning_rate": 1.1961323614771424e-07,
      "loss": 0.1525,
      "num_input_tokens_seen": 62338232,
      "step": 38755
    },
    {
      "epoch": 1.7546571900676793,
      "grad_norm": 0.3541947603225708,
      "learning_rate": 1.1865594302886418e-07,
      "loss": 0.1219,
      "num_input_tokens_seen": 62345912,
      "step": 38760
    },
    {
      "epoch": 1.7548835419543223,
      "grad_norm": 0.33870384097099304,
      "learning_rate": 1.1770248692594687e-07,
      "loss": 0.1539,
      "num_input_tokens_seen": 62353816,
      "step": 38765
    },
    {
      "epoch": 1.7551098938409653,
      "grad_norm": 0.5154322981834412,
      "learning_rate": 1.167528679859975e-07,
      "loss": 0.1338,
      "num_input_tokens_seen": 62361592,
      "step": 38770
    },
    {
      "epoch": 1.7553362457276083,
      "grad_norm": 1.1542519330978394,
      "learning_rate": 1.1580708635545446e-07,
      "loss": 0.1754,
      "num_input_tokens_seen": 62369464,
      "step": 38775
    },
    {
      "epoch": 1.7555625976142513,
      "grad_norm": 0.5759274363517761,
      "learning_rate": 1.1486514218017885e-07,
      "loss": 0.1338,
      "num_input_tokens_seen": 62377592,
      "step": 38780
    },
    {
      "epoch": 1.755788949500894,
      "grad_norm": 0.8136069774627686,
      "learning_rate": 1.1392703560542117e-07,
      "loss": 0.1684,
      "num_input_tokens_seen": 62386008,
      "step": 38785
    },
    {
      "epoch": 1.756015301387537,
      "grad_norm": 0.9175392985343933,
      "learning_rate": 1.129927667758518e-07,
      "loss": 0.1536,
      "num_input_tokens_seen": 62393976,
      "step": 38790
    },
    {
      "epoch": 1.75624165327418,
      "grad_norm": 0.8216807842254639,
      "learning_rate": 1.1206233583554992e-07,
      "loss": 0.1749,
      "num_input_tokens_seen": 62402520,
      "step": 38795
    },
    {
      "epoch": 1.756468005160823,
      "grad_norm": 0.45536184310913086,
      "learning_rate": 1.1113574292799523e-07,
      "loss": 0.1069,
      "num_input_tokens_seen": 62410264,
      "step": 38800
    },
    {
      "epoch": 1.756468005160823,
      "eval_loss": 0.14332839846611023,
      "eval_runtime": 405.5177,
      "eval_samples_per_second": 96.842,
      "eval_steps_per_second": 24.211,
      "num_input_tokens_seen": 62410264,
      "step": 38800
    },
    {
      "epoch": 1.756694357047466,
      "grad_norm": 0.9614572525024414,
      "learning_rate": 1.1021298819608449e-07,
      "loss": 0.1416,
      "num_input_tokens_seen": 62418136,
      "step": 38805
    },
    {
      "epoch": 1.756920708934109,
      "grad_norm": 0.45744815468788147,
      "learning_rate": 1.0929407178211226e-07,
      "loss": 0.1573,
      "num_input_tokens_seen": 62426296,
      "step": 38810
    },
    {
      "epoch": 1.7571470608207518,
      "grad_norm": 0.757989227771759,
      "learning_rate": 1.0837899382779293e-07,
      "loss": 0.1648,
      "num_input_tokens_seen": 62435160,
      "step": 38815
    },
    {
      "epoch": 1.7573734127073948,
      "grad_norm": 0.8066496253013611,
      "learning_rate": 1.0746775447423862e-07,
      "loss": 0.1224,
      "num_input_tokens_seen": 62443064,
      "step": 38820
    },
    {
      "epoch": 1.7575997645940378,
      "grad_norm": 0.6173424124717712,
      "learning_rate": 1.0656035386197583e-07,
      "loss": 0.0966,
      "num_input_tokens_seen": 62451576,
      "step": 38825
    },
    {
      "epoch": 1.7578261164806808,
      "grad_norm": 0.6721802353858948,
      "learning_rate": 1.0565679213093982e-07,
      "loss": 0.125,
      "num_input_tokens_seen": 62459576,
      "step": 38830
    },
    {
      "epoch": 1.7580524683673238,
      "grad_norm": 0.4354250729084015,
      "learning_rate": 1.0475706942046638e-07,
      "loss": 0.1723,
      "num_input_tokens_seen": 62467896,
      "step": 38835
    },
    {
      "epoch": 1.7582788202539668,
      "grad_norm": 0.40838080644607544,
      "learning_rate": 1.0386118586930282e-07,
      "loss": 0.1478,
      "num_input_tokens_seen": 62475928,
      "step": 38840
    },
    {
      "epoch": 1.7585051721406098,
      "grad_norm": 0.5637879967689514,
      "learning_rate": 1.0296914161561367e-07,
      "loss": 0.1352,
      "num_input_tokens_seen": 62484120,
      "step": 38845
    },
    {
      "epoch": 1.7587315240272527,
      "grad_norm": 2.657611131668091,
      "learning_rate": 1.0208093679695552e-07,
      "loss": 0.1494,
      "num_input_tokens_seen": 62492088,
      "step": 38850
    },
    {
      "epoch": 1.7589578759138957,
      "grad_norm": 0.8331004977226257,
      "learning_rate": 1.0119657155030493e-07,
      "loss": 0.1085,
      "num_input_tokens_seen": 62500248,
      "step": 38855
    },
    {
      "epoch": 1.7591842278005387,
      "grad_norm": 0.5484829545021057,
      "learning_rate": 1.003160460120417e-07,
      "loss": 0.1553,
      "num_input_tokens_seen": 62508472,
      "step": 38860
    },
    {
      "epoch": 1.7594105796871817,
      "grad_norm": 0.5244439244270325,
      "learning_rate": 9.943936031795165e-08,
      "loss": 0.1287,
      "num_input_tokens_seen": 62516792,
      "step": 38865
    },
    {
      "epoch": 1.7596369315738247,
      "grad_norm": 0.43489742279052734,
      "learning_rate": 9.856651460323219e-08,
      "loss": 0.1277,
      "num_input_tokens_seen": 62524728,
      "step": 38870
    },
    {
      "epoch": 1.7598632834604677,
      "grad_norm": 1.274726152420044,
      "learning_rate": 9.769750900248953e-08,
      "loss": 0.15,
      "num_input_tokens_seen": 62532600,
      "step": 38875
    },
    {
      "epoch": 1.7600896353471107,
      "grad_norm": 0.756946325302124,
      "learning_rate": 9.683234364973038e-08,
      "loss": 0.1698,
      "num_input_tokens_seen": 62540664,
      "step": 38880
    },
    {
      "epoch": 1.7603159872337537,
      "grad_norm": 0.9026391506195068,
      "learning_rate": 9.597101867837854e-08,
      "loss": 0.1444,
      "num_input_tokens_seen": 62549272,
      "step": 38885
    },
    {
      "epoch": 1.7605423391203967,
      "grad_norm": 0.9342148303985596,
      "learning_rate": 9.511353422125835e-08,
      "loss": 0.1668,
      "num_input_tokens_seen": 62556728,
      "step": 38890
    },
    {
      "epoch": 1.7607686910070397,
      "grad_norm": 1.2861849069595337,
      "learning_rate": 9.42598904106029e-08,
      "loss": 0.1488,
      "num_input_tokens_seen": 62564888,
      "step": 38895
    },
    {
      "epoch": 1.7609950428936825,
      "grad_norm": 0.36429956555366516,
      "learning_rate": 9.341008737806245e-08,
      "loss": 0.1156,
      "num_input_tokens_seen": 62573144,
      "step": 38900
    },
    {
      "epoch": 1.7612213947803255,
      "grad_norm": 0.3899840712547302,
      "learning_rate": 9.256412525467661e-08,
      "loss": 0.1195,
      "num_input_tokens_seen": 62581400,
      "step": 38905
    },
    {
      "epoch": 1.7614477466669685,
      "grad_norm": 0.5414713025093079,
      "learning_rate": 9.172200417091326e-08,
      "loss": 0.1173,
      "num_input_tokens_seen": 62589560,
      "step": 38910
    },
    {
      "epoch": 1.7616740985536115,
      "grad_norm": 1.4149361848831177,
      "learning_rate": 9.088372425663239e-08,
      "loss": 0.1613,
      "num_input_tokens_seen": 62597272,
      "step": 38915
    },
    {
      "epoch": 1.7619004504402545,
      "grad_norm": 0.5038463473320007,
      "learning_rate": 9.004928564110837e-08,
      "loss": 0.091,
      "num_input_tokens_seen": 62605464,
      "step": 38920
    },
    {
      "epoch": 1.7621268023268974,
      "grad_norm": 0.5460103154182434,
      "learning_rate": 8.92186884530244e-08,
      "loss": 0.1783,
      "num_input_tokens_seen": 62613176,
      "step": 38925
    },
    {
      "epoch": 1.7623531542135402,
      "grad_norm": 0.4014289975166321,
      "learning_rate": 8.83919328204641e-08,
      "loss": 0.109,
      "num_input_tokens_seen": 62621464,
      "step": 38930
    },
    {
      "epoch": 1.7625795061001832,
      "grad_norm": 0.598773181438446,
      "learning_rate": 8.756901887093105e-08,
      "loss": 0.1404,
      "num_input_tokens_seen": 62629432,
      "step": 38935
    },
    {
      "epoch": 1.7628058579868262,
      "grad_norm": 0.8757213354110718,
      "learning_rate": 8.674994673132098e-08,
      "loss": 0.1841,
      "num_input_tokens_seen": 62637240,
      "step": 38940
    },
    {
      "epoch": 1.7630322098734692,
      "grad_norm": 0.40528127551078796,
      "learning_rate": 8.593471652794949e-08,
      "loss": 0.1179,
      "num_input_tokens_seen": 62645112,
      "step": 38945
    },
    {
      "epoch": 1.7632585617601122,
      "grad_norm": 1.3024332523345947,
      "learning_rate": 8.512332838653548e-08,
      "loss": 0.1442,
      "num_input_tokens_seen": 62653048,
      "step": 38950
    },
    {
      "epoch": 1.7634849136467552,
      "grad_norm": 0.7866619825363159,
      "learning_rate": 8.431578243220106e-08,
      "loss": 0.1232,
      "num_input_tokens_seen": 62662040,
      "step": 38955
    },
    {
      "epoch": 1.7637112655333982,
      "grad_norm": 1.3959996700286865,
      "learning_rate": 8.351207878948552e-08,
      "loss": 0.1444,
      "num_input_tokens_seen": 62669880,
      "step": 38960
    },
    {
      "epoch": 1.7639376174200412,
      "grad_norm": 0.7112109065055847,
      "learning_rate": 8.271221758232583e-08,
      "loss": 0.152,
      "num_input_tokens_seen": 62677752,
      "step": 38965
    },
    {
      "epoch": 1.7641639693066842,
      "grad_norm": 0.8558568358421326,
      "learning_rate": 8.191619893407332e-08,
      "loss": 0.1842,
      "num_input_tokens_seen": 62685688,
      "step": 38970
    },
    {
      "epoch": 1.7643903211933272,
      "grad_norm": 1.0448622703552246,
      "learning_rate": 8.112402296748534e-08,
      "loss": 0.118,
      "num_input_tokens_seen": 62693688,
      "step": 38975
    },
    {
      "epoch": 1.7646166730799702,
      "grad_norm": 0.8129662871360779,
      "learning_rate": 8.033568980471973e-08,
      "loss": 0.1424,
      "num_input_tokens_seen": 62701272,
      "step": 38980
    },
    {
      "epoch": 1.7648430249666132,
      "grad_norm": 0.7020276784896851,
      "learning_rate": 7.955119956735146e-08,
      "loss": 0.1247,
      "num_input_tokens_seen": 62709176,
      "step": 38985
    },
    {
      "epoch": 1.7650693768532562,
      "grad_norm": 0.5721438527107239,
      "learning_rate": 7.877055237636155e-08,
      "loss": 0.1833,
      "num_input_tokens_seen": 62717208,
      "step": 38990
    },
    {
      "epoch": 1.7652957287398992,
      "grad_norm": 1.0944209098815918,
      "learning_rate": 7.79937483521287e-08,
      "loss": 0.126,
      "num_input_tokens_seen": 62724984,
      "step": 38995
    },
    {
      "epoch": 1.7655220806265421,
      "grad_norm": 0.7274468541145325,
      "learning_rate": 7.722078761444873e-08,
      "loss": 0.1101,
      "num_input_tokens_seen": 62732824,
      "step": 39000
    },
    {
      "epoch": 1.7655220806265421,
      "eval_loss": 0.14328044652938843,
      "eval_runtime": 405.1862,
      "eval_samples_per_second": 96.921,
      "eval_steps_per_second": 24.231,
      "num_input_tokens_seen": 62732824,
      "step": 39000
    },
    {
      "epoch": 1.7657484325131851,
      "grad_norm": 0.47750377655029297,
      "learning_rate": 7.645167028252631e-08,
      "loss": 0.1316,
      "num_input_tokens_seen": 62740728,
      "step": 39005
    },
    {
      "epoch": 1.7659747843998281,
      "grad_norm": 0.8020319938659668,
      "learning_rate": 7.568639647496379e-08,
      "loss": 0.1833,
      "num_input_tokens_seen": 62748888,
      "step": 39010
    },
    {
      "epoch": 1.766201136286471,
      "grad_norm": 0.45891639590263367,
      "learning_rate": 7.492496630977508e-08,
      "loss": 0.1382,
      "num_input_tokens_seen": 62756760,
      "step": 39015
    },
    {
      "epoch": 1.766427488173114,
      "grad_norm": 0.5373697876930237,
      "learning_rate": 7.416737990438571e-08,
      "loss": 0.1374,
      "num_input_tokens_seen": 62764856,
      "step": 39020
    },
    {
      "epoch": 1.766653840059757,
      "grad_norm": 0.9441350102424622,
      "learning_rate": 7.341363737562445e-08,
      "loss": 0.1458,
      "num_input_tokens_seen": 62773080,
      "step": 39025
    },
    {
      "epoch": 1.7668801919464,
      "grad_norm": 0.45341572165489197,
      "learning_rate": 7.266373883972887e-08,
      "loss": 0.0942,
      "num_input_tokens_seen": 62781304,
      "step": 39030
    },
    {
      "epoch": 1.7671065438330429,
      "grad_norm": 0.3450321853160858,
      "learning_rate": 7.191768441233981e-08,
      "loss": 0.1207,
      "num_input_tokens_seen": 62789016,
      "step": 39035
    },
    {
      "epoch": 1.7673328957196857,
      "grad_norm": 0.9479008316993713,
      "learning_rate": 7.11754742085069e-08,
      "loss": 0.1108,
      "num_input_tokens_seen": 62797688,
      "step": 39040
    },
    {
      "epoch": 1.7675592476063287,
      "grad_norm": 0.792230486869812,
      "learning_rate": 7.043710834269413e-08,
      "loss": 0.127,
      "num_input_tokens_seen": 62805592,
      "step": 39045
    },
    {
      "epoch": 1.7677855994929716,
      "grad_norm": 0.836820125579834,
      "learning_rate": 6.970258692876319e-08,
      "loss": 0.1658,
      "num_input_tokens_seen": 62812888,
      "step": 39050
    },
    {
      "epoch": 1.7680119513796146,
      "grad_norm": 0.8854153752326965,
      "learning_rate": 6.897191007998738e-08,
      "loss": 0.1679,
      "num_input_tokens_seen": 62820696,
      "step": 39055
    },
    {
      "epoch": 1.7682383032662576,
      "grad_norm": 0.689755916595459,
      "learning_rate": 6.824507790904599e-08,
      "loss": 0.1184,
      "num_input_tokens_seen": 62828024,
      "step": 39060
    },
    {
      "epoch": 1.7684646551529006,
      "grad_norm": 0.5517419576644897,
      "learning_rate": 6.752209052802439e-08,
      "loss": 0.1499,
      "num_input_tokens_seen": 62836536,
      "step": 39065
    },
    {
      "epoch": 1.7686910070395436,
      "grad_norm": 0.48115459084510803,
      "learning_rate": 6.680294804841946e-08,
      "loss": 0.1543,
      "num_input_tokens_seen": 62843896,
      "step": 39070
    },
    {
      "epoch": 1.7689173589261866,
      "grad_norm": 0.39687275886535645,
      "learning_rate": 6.608765058112865e-08,
      "loss": 0.1368,
      "num_input_tokens_seen": 62851896,
      "step": 39075
    },
    {
      "epoch": 1.7691437108128296,
      "grad_norm": 0.5759522318840027,
      "learning_rate": 6.537619823646368e-08,
      "loss": 0.1247,
      "num_input_tokens_seen": 62859960,
      "step": 39080
    },
    {
      "epoch": 1.7693700626994726,
      "grad_norm": 0.5194347500801086,
      "learning_rate": 6.466859112413404e-08,
      "loss": 0.1253,
      "num_input_tokens_seen": 62868184,
      "step": 39085
    },
    {
      "epoch": 1.7695964145861156,
      "grad_norm": 0.8272833824157715,
      "learning_rate": 6.39648293532663e-08,
      "loss": 0.148,
      "num_input_tokens_seen": 62876312,
      "step": 39090
    },
    {
      "epoch": 1.7698227664727586,
      "grad_norm": 1.268498420715332,
      "learning_rate": 6.32649130323848e-08,
      "loss": 0.1231,
      "num_input_tokens_seen": 62884792,
      "step": 39095
    },
    {
      "epoch": 1.7700491183594016,
      "grad_norm": 0.7414143085479736,
      "learning_rate": 6.256884226943094e-08,
      "loss": 0.1773,
      "num_input_tokens_seen": 62892920,
      "step": 39100
    },
    {
      "epoch": 1.7702754702460446,
      "grad_norm": 0.5286200046539307,
      "learning_rate": 6.187661717174386e-08,
      "loss": 0.1493,
      "num_input_tokens_seen": 62902072,
      "step": 39105
    },
    {
      "epoch": 1.7705018221326876,
      "grad_norm": 0.6687130331993103,
      "learning_rate": 6.118823784607708e-08,
      "loss": 0.1546,
      "num_input_tokens_seen": 62909784,
      "step": 39110
    },
    {
      "epoch": 1.7707281740193306,
      "grad_norm": 0.2951086163520813,
      "learning_rate": 6.050370439858178e-08,
      "loss": 0.1282,
      "num_input_tokens_seen": 62918584,
      "step": 39115
    },
    {
      "epoch": 1.7709545259059736,
      "grad_norm": 0.3790076971054077,
      "learning_rate": 5.98230169348235e-08,
      "loss": 0.1474,
      "num_input_tokens_seen": 62927128,
      "step": 39120
    },
    {
      "epoch": 1.7711808777926163,
      "grad_norm": 1.0147786140441895,
      "learning_rate": 5.914617555977664e-08,
      "loss": 0.1305,
      "num_input_tokens_seen": 62934936,
      "step": 39125
    },
    {
      "epoch": 1.7714072296792593,
      "grad_norm": 0.4037209749221802,
      "learning_rate": 5.8473180377816017e-08,
      "loss": 0.1353,
      "num_input_tokens_seen": 62942840,
      "step": 39130
    },
    {
      "epoch": 1.7716335815659023,
      "grad_norm": 0.9783034920692444,
      "learning_rate": 5.780403149272251e-08,
      "loss": 0.2003,
      "num_input_tokens_seen": 62951448,
      "step": 39135
    },
    {
      "epoch": 1.7718599334525453,
      "grad_norm": 0.6874833703041077,
      "learning_rate": 5.7138729007694126e-08,
      "loss": 0.114,
      "num_input_tokens_seen": 62959064,
      "step": 39140
    },
    {
      "epoch": 1.7720862853391883,
      "grad_norm": 0.5166005492210388,
      "learning_rate": 5.64772730253238e-08,
      "loss": 0.1308,
      "num_input_tokens_seen": 62967448,
      "step": 39145
    },
    {
      "epoch": 1.7723126372258313,
      "grad_norm": 0.3944481611251831,
      "learning_rate": 5.5819663647618814e-08,
      "loss": 0.1556,
      "num_input_tokens_seen": 62975576,
      "step": 39150
    },
    {
      "epoch": 1.772538989112474,
      "grad_norm": 1.1136418581008911,
      "learning_rate": 5.5165900975989723e-08,
      "loss": 0.1635,
      "num_input_tokens_seen": 62982680,
      "step": 39155
    },
    {
      "epoch": 1.772765340999117,
      "grad_norm": 0.3824903070926666,
      "learning_rate": 5.451598511125311e-08,
      "loss": 0.1325,
      "num_input_tokens_seen": 62990968,
      "step": 39160
    },
    {
      "epoch": 1.77299169288576,
      "grad_norm": 0.6201068758964539,
      "learning_rate": 5.3869916153637124e-08,
      "loss": 0.1734,
      "num_input_tokens_seen": 62998776,
      "step": 39165
    },
    {
      "epoch": 1.773218044772403,
      "grad_norm": 0.8548918962478638,
      "learning_rate": 5.322769420277318e-08,
      "loss": 0.1682,
      "num_input_tokens_seen": 63007192,
      "step": 39170
    },
    {
      "epoch": 1.773444396659046,
      "grad_norm": 0.7841001152992249,
      "learning_rate": 5.258931935769873e-08,
      "loss": 0.113,
      "num_input_tokens_seen": 63015160,
      "step": 39175
    },
    {
      "epoch": 1.773670748545689,
      "grad_norm": 1.0107171535491943,
      "learning_rate": 5.19547917168628e-08,
      "loss": 0.2004,
      "num_input_tokens_seen": 63022744,
      "step": 39180
    },
    {
      "epoch": 1.773897100432332,
      "grad_norm": 0.5022920370101929,
      "learning_rate": 5.13241113781121e-08,
      "loss": 0.1469,
      "num_input_tokens_seen": 63030584,
      "step": 39185
    },
    {
      "epoch": 1.774123452318975,
      "grad_norm": 1.1653690338134766,
      "learning_rate": 5.0697278438707755e-08,
      "loss": 0.1661,
      "num_input_tokens_seen": 63038520,
      "step": 39190
    },
    {
      "epoch": 1.774349804205618,
      "grad_norm": 0.3593589663505554,
      "learning_rate": 5.0074292995316854e-08,
      "loss": 0.1155,
      "num_input_tokens_seen": 63046072,
      "step": 39195
    },
    {
      "epoch": 1.774576156092261,
      "grad_norm": 0.7059395909309387,
      "learning_rate": 4.945515514400978e-08,
      "loss": 0.1397,
      "num_input_tokens_seen": 63054168,
      "step": 39200
    },
    {
      "epoch": 1.774576156092261,
      "eval_loss": 0.1431099772453308,
      "eval_runtime": 404.1322,
      "eval_samples_per_second": 97.174,
      "eval_steps_per_second": 24.294,
      "num_input_tokens_seen": 63054168,
      "step": 39200
    },
    {
      "epoch": 1.774802507978904,
      "grad_norm": 0.32426637411117554,
      "learning_rate": 4.883986498026571e-08,
      "loss": 0.155,
      "num_input_tokens_seen": 63062776,
      "step": 39205
    },
    {
      "epoch": 1.775028859865547,
      "grad_norm": 0.7025501132011414,
      "learning_rate": 4.822842259896987e-08,
      "loss": 0.1393,
      "num_input_tokens_seen": 63070872,
      "step": 39210
    },
    {
      "epoch": 1.77525521175219,
      "grad_norm": 0.41098690032958984,
      "learning_rate": 4.762082809441626e-08,
      "loss": 0.1312,
      "num_input_tokens_seen": 63078872,
      "step": 39215
    },
    {
      "epoch": 1.775481563638833,
      "grad_norm": 0.8670307397842407,
      "learning_rate": 4.7017081560302156e-08,
      "loss": 0.1126,
      "num_input_tokens_seen": 63087192,
      "step": 39220
    },
    {
      "epoch": 1.775707915525476,
      "grad_norm": 0.3535425364971161,
      "learning_rate": 4.6417183089730866e-08,
      "loss": 0.1187,
      "num_input_tokens_seen": 63095256,
      "step": 39225
    },
    {
      "epoch": 1.775934267412119,
      "grad_norm": 1.0761562585830688,
      "learning_rate": 4.5821132775217265e-08,
      "loss": 0.1073,
      "num_input_tokens_seen": 63103160,
      "step": 39230
    },
    {
      "epoch": 1.776160619298762,
      "grad_norm": 0.9014934301376343,
      "learning_rate": 4.5228930708679504e-08,
      "loss": 0.1965,
      "num_input_tokens_seen": 63111032,
      "step": 39235
    },
    {
      "epoch": 1.7763869711854048,
      "grad_norm": 0.9050735235214233,
      "learning_rate": 4.464057698144175e-08,
      "loss": 0.1561,
      "num_input_tokens_seen": 63118808,
      "step": 39240
    },
    {
      "epoch": 1.7766133230720478,
      "grad_norm": 0.5585442185401917,
      "learning_rate": 4.4056071684236974e-08,
      "loss": 0.1136,
      "num_input_tokens_seen": 63127416,
      "step": 39245
    },
    {
      "epoch": 1.7768396749586908,
      "grad_norm": 0.738108217716217,
      "learning_rate": 4.347541490719864e-08,
      "loss": 0.1189,
      "num_input_tokens_seen": 63136280,
      "step": 39250
    },
    {
      "epoch": 1.7770660268453338,
      "grad_norm": 1.372745394706726,
      "learning_rate": 4.2898606739877336e-08,
      "loss": 0.1494,
      "num_input_tokens_seen": 63144312,
      "step": 39255
    },
    {
      "epoch": 1.7772923787319768,
      "grad_norm": 0.7828232049942017,
      "learning_rate": 4.232564727122135e-08,
      "loss": 0.108,
      "num_input_tokens_seen": 63152056,
      "step": 39260
    },
    {
      "epoch": 1.7775187306186198,
      "grad_norm": 0.4192749857902527,
      "learning_rate": 4.1756536589585004e-08,
      "loss": 0.1476,
      "num_input_tokens_seen": 63160088,
      "step": 39265
    },
    {
      "epoch": 1.7777450825052625,
      "grad_norm": 0.4262440502643585,
      "learning_rate": 4.119127478273976e-08,
      "loss": 0.1376,
      "num_input_tokens_seen": 63167576,
      "step": 39270
    },
    {
      "epoch": 1.7779714343919055,
      "grad_norm": 0.2605864107608795,
      "learning_rate": 4.062986193784923e-08,
      "loss": 0.1295,
      "num_input_tokens_seen": 63175224,
      "step": 39275
    },
    {
      "epoch": 1.7781977862785485,
      "grad_norm": 0.8065060973167419,
      "learning_rate": 4.007229814149416e-08,
      "loss": 0.1928,
      "num_input_tokens_seen": 63183672,
      "step": 39280
    },
    {
      "epoch": 1.7784241381651915,
      "grad_norm": 1.0438873767852783,
      "learning_rate": 3.951858347965576e-08,
      "loss": 0.1219,
      "num_input_tokens_seen": 63191704,
      "step": 39285
    },
    {
      "epoch": 1.7786504900518345,
      "grad_norm": 0.4607962667942047,
      "learning_rate": 3.896871803772684e-08,
      "loss": 0.1323,
      "num_input_tokens_seen": 63199992,
      "step": 39290
    },
    {
      "epoch": 1.7788768419384775,
      "grad_norm": 0.34493720531463623,
      "learning_rate": 3.842270190050068e-08,
      "loss": 0.1253,
      "num_input_tokens_seen": 63208120,
      "step": 39295
    },
    {
      "epoch": 1.7791031938251205,
      "grad_norm": 0.7874628305435181,
      "learning_rate": 3.7880535152179376e-08,
      "loss": 0.0979,
      "num_input_tokens_seen": 63216280,
      "step": 39300
    },
    {
      "epoch": 1.7793295457117635,
      "grad_norm": 1.105746865272522,
      "learning_rate": 3.734221787637382e-08,
      "loss": 0.1391,
      "num_input_tokens_seen": 63224184,
      "step": 39305
    },
    {
      "epoch": 1.7795558975984065,
      "grad_norm": 0.42733097076416016,
      "learning_rate": 3.680775015609817e-08,
      "loss": 0.1475,
      "num_input_tokens_seen": 63231992,
      "step": 39310
    },
    {
      "epoch": 1.7797822494850495,
      "grad_norm": 0.4200257956981659,
      "learning_rate": 3.627713207377537e-08,
      "loss": 0.0995,
      "num_input_tokens_seen": 63239800,
      "step": 39315
    },
    {
      "epoch": 1.7800086013716925,
      "grad_norm": 0.38127291202545166,
      "learning_rate": 3.575036371123164e-08,
      "loss": 0.1216,
      "num_input_tokens_seen": 63248088,
      "step": 39320
    },
    {
      "epoch": 1.7802349532583355,
      "grad_norm": 0.6058573126792908,
      "learning_rate": 3.5227445149704776e-08,
      "loss": 0.1283,
      "num_input_tokens_seen": 63256248,
      "step": 39325
    },
    {
      "epoch": 1.7804613051449785,
      "grad_norm": 1.1427947282791138,
      "learning_rate": 3.470837646983027e-08,
      "loss": 0.1052,
      "num_input_tokens_seen": 63264248,
      "step": 39330
    },
    {
      "epoch": 1.7806876570316215,
      "grad_norm": 0.4872131943702698,
      "learning_rate": 3.419315775165799e-08,
      "loss": 0.1119,
      "num_input_tokens_seen": 63272152,
      "step": 39335
    },
    {
      "epoch": 1.7809140089182645,
      "grad_norm": 0.5111562609672546,
      "learning_rate": 3.368178907464103e-08,
      "loss": 0.1309,
      "num_input_tokens_seen": 63279992,
      "step": 39340
    },
    {
      "epoch": 1.7811403608049075,
      "grad_norm": 0.5129215717315674,
      "learning_rate": 3.317427051763855e-08,
      "loss": 0.1306,
      "num_input_tokens_seen": 63287928,
      "step": 39345
    },
    {
      "epoch": 1.7813667126915504,
      "grad_norm": 0.5851658582687378,
      "learning_rate": 3.267060215891571e-08,
      "loss": 0.181,
      "num_input_tokens_seen": 63295768,
      "step": 39350
    },
    {
      "epoch": 1.7815930645781932,
      "grad_norm": 0.5770663619041443,
      "learning_rate": 3.217078407614649e-08,
      "loss": 0.1328,
      "num_input_tokens_seen": 63303672,
      "step": 39355
    },
    {
      "epoch": 1.7818194164648362,
      "grad_norm": 1.6752772331237793,
      "learning_rate": 3.1674816346405345e-08,
      "loss": 0.1387,
      "num_input_tokens_seen": 63311704,
      "step": 39360
    },
    {
      "epoch": 1.7820457683514792,
      "grad_norm": 0.3849746882915497,
      "learning_rate": 3.11826990461811e-08,
      "loss": 0.0777,
      "num_input_tokens_seen": 63319800,
      "step": 39365
    },
    {
      "epoch": 1.7822721202381222,
      "grad_norm": 0.6239516139030457,
      "learning_rate": 3.069443225136304e-08,
      "loss": 0.151,
      "num_input_tokens_seen": 63327896,
      "step": 39370
    },
    {
      "epoch": 1.7824984721247652,
      "grad_norm": 0.7493324279785156,
      "learning_rate": 3.021001603724372e-08,
      "loss": 0.1365,
      "num_input_tokens_seen": 63335512,
      "step": 39375
    },
    {
      "epoch": 1.782724824011408,
      "grad_norm": 0.6843831539154053,
      "learning_rate": 2.9729450478532818e-08,
      "loss": 0.1011,
      "num_input_tokens_seen": 63343512,
      "step": 39380
    },
    {
      "epoch": 1.782951175898051,
      "grad_norm": 0.5888940095901489,
      "learning_rate": 2.9252735649337726e-08,
      "loss": 0.1562,
      "num_input_tokens_seen": 63351928,
      "step": 39385
    },
    {
      "epoch": 1.783177527784694,
      "grad_norm": 1.0032466650009155,
      "learning_rate": 2.8779871623171863e-08,
      "loss": 0.1265,
      "num_input_tokens_seen": 63359576,
      "step": 39390
    },
    {
      "epoch": 1.783403879671337,
      "grad_norm": 0.5493446588516235,
      "learning_rate": 2.8310858472957448e-08,
      "loss": 0.1098,
      "num_input_tokens_seen": 63367480,
      "step": 39395
    },
    {
      "epoch": 1.78363023155798,
      "grad_norm": 0.6458383202552795,
      "learning_rate": 2.784569627101996e-08,
      "loss": 0.1819,
      "num_input_tokens_seen": 63375640,
      "step": 39400
    },
    {
      "epoch": 1.78363023155798,
      "eval_loss": 0.14331088960170746,
      "eval_runtime": 405.4123,
      "eval_samples_per_second": 96.867,
      "eval_steps_per_second": 24.217,
      "num_input_tokens_seen": 63375640,
      "step": 39400
    },
    {
      "epoch": 1.783856583444623,
      "grad_norm": 0.6600026488304138,
      "learning_rate": 2.738438508909924e-08,
      "loss": 0.1674,
      "num_input_tokens_seen": 63383224,
      "step": 39405
    },
    {
      "epoch": 1.784082935331266,
      "grad_norm": 0.8793392777442932,
      "learning_rate": 2.692692499833005e-08,
      "loss": 0.1567,
      "num_input_tokens_seen": 63391480,
      "step": 39410
    },
    {
      "epoch": 1.784309287217909,
      "grad_norm": 0.4471677541732788,
      "learning_rate": 2.647331606926151e-08,
      "loss": 0.1318,
      "num_input_tokens_seen": 63399256,
      "step": 39415
    },
    {
      "epoch": 1.784535639104552,
      "grad_norm": 0.7660313248634338,
      "learning_rate": 2.6023558371843225e-08,
      "loss": 0.1843,
      "num_input_tokens_seen": 63407096,
      "step": 39420
    },
    {
      "epoch": 1.784761990991195,
      "grad_norm": 0.5558164119720459,
      "learning_rate": 2.557765197543638e-08,
      "loss": 0.149,
      "num_input_tokens_seen": 63414872,
      "step": 39425
    },
    {
      "epoch": 1.784988342877838,
      "grad_norm": 0.8069942593574524,
      "learning_rate": 2.513559694880263e-08,
      "loss": 0.0949,
      "num_input_tokens_seen": 63422808,
      "step": 39430
    },
    {
      "epoch": 1.785214694764481,
      "grad_norm": 0.9343995451927185,
      "learning_rate": 2.469739336011523e-08,
      "loss": 0.1516,
      "num_input_tokens_seen": 63430840,
      "step": 39435
    },
    {
      "epoch": 1.785441046651124,
      "grad_norm": 0.3024303615093231,
      "learning_rate": 2.4263041276947894e-08,
      "loss": 0.1093,
      "num_input_tokens_seen": 63438520,
      "step": 39440
    },
    {
      "epoch": 1.785667398537767,
      "grad_norm": 0.27146485447883606,
      "learning_rate": 2.3832540766283164e-08,
      "loss": 0.1168,
      "num_input_tokens_seen": 63446456,
      "step": 39445
    },
    {
      "epoch": 1.78589375042441,
      "grad_norm": 0.7677844166755676,
      "learning_rate": 2.3405891894512366e-08,
      "loss": 0.1543,
      "num_input_tokens_seen": 63454584,
      "step": 39450
    },
    {
      "epoch": 1.786120102311053,
      "grad_norm": 0.45113685727119446,
      "learning_rate": 2.29830947274301e-08,
      "loss": 0.1333,
      "num_input_tokens_seen": 63462776,
      "step": 39455
    },
    {
      "epoch": 1.7863464541976959,
      "grad_norm": 0.7892574667930603,
      "learning_rate": 2.2564149330231432e-08,
      "loss": 0.1563,
      "num_input_tokens_seen": 63471000,
      "step": 39460
    },
    {
      "epoch": 1.7865728060843387,
      "grad_norm": 0.5270522236824036,
      "learning_rate": 2.2149055767528572e-08,
      "loss": 0.1383,
      "num_input_tokens_seen": 63478936,
      "step": 39465
    },
    {
      "epoch": 1.7867991579709817,
      "grad_norm": 0.5624051690101624,
      "learning_rate": 2.1737814103334197e-08,
      "loss": 0.1397,
      "num_input_tokens_seen": 63487480,
      "step": 39470
    },
    {
      "epoch": 1.7870255098576246,
      "grad_norm": 0.9233453869819641,
      "learning_rate": 2.1330424401064253e-08,
      "loss": 0.1405,
      "num_input_tokens_seen": 63495384,
      "step": 39475
    },
    {
      "epoch": 1.7872518617442676,
      "grad_norm": 0.6809916496276855,
      "learning_rate": 2.092688672354348e-08,
      "loss": 0.1376,
      "num_input_tokens_seen": 63503576,
      "step": 39480
    },
    {
      "epoch": 1.7874782136309106,
      "grad_norm": 1.0571528673171997,
      "learning_rate": 2.0527201133005435e-08,
      "loss": 0.1406,
      "num_input_tokens_seen": 63511768,
      "step": 39485
    },
    {
      "epoch": 1.7877045655175536,
      "grad_norm": 0.2925744652748108,
      "learning_rate": 2.0131367691084148e-08,
      "loss": 0.1143,
      "num_input_tokens_seen": 63520312,
      "step": 39490
    },
    {
      "epoch": 1.7879309174041964,
      "grad_norm": 0.5680410265922546,
      "learning_rate": 1.9739386458819675e-08,
      "loss": 0.1223,
      "num_input_tokens_seen": 63528472,
      "step": 39495
    },
    {
      "epoch": 1.7881572692908394,
      "grad_norm": 1.2707440853118896,
      "learning_rate": 1.9351257496666442e-08,
      "loss": 0.1749,
      "num_input_tokens_seen": 63536152,
      "step": 39500
    },
    {
      "epoch": 1.7883836211774824,
      "grad_norm": 0.6917542815208435,
      "learning_rate": 1.896698086447657e-08,
      "loss": 0.1512,
      "num_input_tokens_seen": 63544312,
      "step": 39505
    },
    {
      "epoch": 1.7886099730641254,
      "grad_norm": 0.6469237804412842,
      "learning_rate": 1.8586556621505436e-08,
      "loss": 0.1651,
      "num_input_tokens_seen": 63552024,
      "step": 39510
    },
    {
      "epoch": 1.7888363249507684,
      "grad_norm": 0.8127576112747192,
      "learning_rate": 1.820998482642833e-08,
      "loss": 0.1779,
      "num_input_tokens_seen": 63560888,
      "step": 39515
    },
    {
      "epoch": 1.7890626768374114,
      "grad_norm": 0.739342212677002,
      "learning_rate": 1.7837265537309912e-08,
      "loss": 0.1037,
      "num_input_tokens_seen": 63569432,
      "step": 39520
    },
    {
      "epoch": 1.7892890287240544,
      "grad_norm": 0.41831937432289124,
      "learning_rate": 1.7468398811629206e-08,
      "loss": 0.1623,
      "num_input_tokens_seen": 63577176,
      "step": 39525
    },
    {
      "epoch": 1.7895153806106974,
      "grad_norm": 0.39841675758361816,
      "learning_rate": 1.710338470627404e-08,
      "loss": 0.1312,
      "num_input_tokens_seen": 63585176,
      "step": 39530
    },
    {
      "epoch": 1.7897417324973404,
      "grad_norm": 0.45966073870658875,
      "learning_rate": 1.6742223277529945e-08,
      "loss": 0.1465,
      "num_input_tokens_seen": 63592888,
      "step": 39535
    },
    {
      "epoch": 1.7899680843839834,
      "grad_norm": 0.7297216653823853,
      "learning_rate": 1.6384914581094036e-08,
      "loss": 0.1471,
      "num_input_tokens_seen": 63600664,
      "step": 39540
    },
    {
      "epoch": 1.7901944362706264,
      "grad_norm": 1.1742100715637207,
      "learning_rate": 1.6031458672069455e-08,
      "loss": 0.1714,
      "num_input_tokens_seen": 63610456,
      "step": 39545
    },
    {
      "epoch": 1.7904207881572693,
      "grad_norm": 0.6185680031776428,
      "learning_rate": 1.5681855604962602e-08,
      "loss": 0.1177,
      "num_input_tokens_seen": 63618424,
      "step": 39550
    },
    {
      "epoch": 1.7906471400439123,
      "grad_norm": 0.4847458600997925,
      "learning_rate": 1.5336105433683135e-08,
      "loss": 0.1588,
      "num_input_tokens_seen": 63626808,
      "step": 39555
    },
    {
      "epoch": 1.7908734919305553,
      "grad_norm": 0.9852573275566101,
      "learning_rate": 1.499420821155506e-08,
      "loss": 0.149,
      "num_input_tokens_seen": 63634424,
      "step": 39560
    },
    {
      "epoch": 1.7910998438171983,
      "grad_norm": 1.4293874502182007,
      "learning_rate": 1.4656163991302874e-08,
      "loss": 0.1371,
      "num_input_tokens_seen": 63642360,
      "step": 39565
    },
    {
      "epoch": 1.7913261957038413,
      "grad_norm": 0.8450164198875427,
      "learning_rate": 1.4321972825051544e-08,
      "loss": 0.1662,
      "num_input_tokens_seen": 63650456,
      "step": 39570
    },
    {
      "epoch": 1.7915525475904843,
      "grad_norm": 0.34692713618278503,
      "learning_rate": 1.3991634764345951e-08,
      "loss": 0.1659,
      "num_input_tokens_seen": 63658488,
      "step": 39575
    },
    {
      "epoch": 1.791778899477127,
      "grad_norm": 0.5187107920646667,
      "learning_rate": 1.3665149860120352e-08,
      "loss": 0.1229,
      "num_input_tokens_seen": 63666552,
      "step": 39580
    },
    {
      "epoch": 1.79200525136377,
      "grad_norm": 0.48802492022514343,
      "learning_rate": 1.3342518162728912e-08,
      "loss": 0.1501,
      "num_input_tokens_seen": 63674232,
      "step": 39585
    },
    {
      "epoch": 1.792231603250413,
      "grad_norm": 0.5327977538108826,
      "learning_rate": 1.30237397219235e-08,
      "loss": 0.1466,
      "num_input_tokens_seen": 63681944,
      "step": 39590
    },
    {
      "epoch": 1.792457955137056,
      "grad_norm": 0.3580634295940399,
      "learning_rate": 1.2708814586862016e-08,
      "loss": 0.156,
      "num_input_tokens_seen": 63689720,
      "step": 39595
    },
    {
      "epoch": 1.792684307023699,
      "grad_norm": 0.6794991493225098,
      "learning_rate": 1.2397742806111168e-08,
      "loss": 0.1069,
      "num_input_tokens_seen": 63697560,
      "step": 39600
    },
    {
      "epoch": 1.792684307023699,
      "eval_loss": 0.14338818192481995,
      "eval_runtime": 404.9817,
      "eval_samples_per_second": 96.97,
      "eval_steps_per_second": 24.243,
      "num_input_tokens_seen": 63697560,
      "step": 39600
    },
    {
      "epoch": 1.792910658910342,
      "grad_norm": 0.6121728420257568,
      "learning_rate": 1.209052442764369e-08,
      "loss": 0.1821,
      "num_input_tokens_seen": 63705528,
      "step": 39605
    },
    {
      "epoch": 1.7931370107969848,
      "grad_norm": 1.2035279273986816,
      "learning_rate": 1.17871594988328e-08,
      "loss": 0.1534,
      "num_input_tokens_seen": 63713432,
      "step": 39610
    },
    {
      "epoch": 1.7933633626836278,
      "grad_norm": 0.3951411545276642,
      "learning_rate": 1.1487648066466072e-08,
      "loss": 0.1402,
      "num_input_tokens_seen": 63721336,
      "step": 39615
    },
    {
      "epoch": 1.7935897145702708,
      "grad_norm": 0.41666868329048157,
      "learning_rate": 1.1191990176728784e-08,
      "loss": 0.1043,
      "num_input_tokens_seen": 63729880,
      "step": 39620
    },
    {
      "epoch": 1.7938160664569138,
      "grad_norm": 0.30426570773124695,
      "learning_rate": 1.0900185875215018e-08,
      "loss": 0.113,
      "num_input_tokens_seen": 63737816,
      "step": 39625
    },
    {
      "epoch": 1.7940424183435568,
      "grad_norm": 1.4306293725967407,
      "learning_rate": 1.0612235206924891e-08,
      "loss": 0.1574,
      "num_input_tokens_seen": 63745528,
      "step": 39630
    },
    {
      "epoch": 1.7942687702301998,
      "grad_norm": 0.40718960762023926,
      "learning_rate": 1.0328138216264549e-08,
      "loss": 0.1207,
      "num_input_tokens_seen": 63753976,
      "step": 39635
    },
    {
      "epoch": 1.7944951221168428,
      "grad_norm": 0.6355692744255066,
      "learning_rate": 1.004789494704339e-08,
      "loss": 0.1486,
      "num_input_tokens_seen": 63761464,
      "step": 39640
    },
    {
      "epoch": 1.7947214740034858,
      "grad_norm": 0.7347559332847595,
      "learning_rate": 9.771505442482397e-09,
      "loss": 0.124,
      "num_input_tokens_seen": 63769048,
      "step": 39645
    },
    {
      "epoch": 1.7949478258901288,
      "grad_norm": 1.1305122375488281,
      "learning_rate": 9.498969745200259e-09,
      "loss": 0.1395,
      "num_input_tokens_seen": 63777176,
      "step": 39650
    },
    {
      "epoch": 1.7951741777767718,
      "grad_norm": 0.43100807070732117,
      "learning_rate": 9.230287897230017e-09,
      "loss": 0.1195,
      "num_input_tokens_seen": 63784568,
      "step": 39655
    },
    {
      "epoch": 1.7954005296634148,
      "grad_norm": 0.8680978417396545,
      "learning_rate": 8.965459940002419e-09,
      "loss": 0.1822,
      "num_input_tokens_seen": 63792440,
      "step": 39660
    },
    {
      "epoch": 1.7956268815500578,
      "grad_norm": 0.6864256262779236,
      "learning_rate": 8.704485914357019e-09,
      "loss": 0.1806,
      "num_input_tokens_seen": 63801112,
      "step": 39665
    },
    {
      "epoch": 1.7958532334367008,
      "grad_norm": 0.8331483006477356,
      "learning_rate": 8.447365860539402e-09,
      "loss": 0.1243,
      "num_input_tokens_seen": 63809240,
      "step": 39670
    },
    {
      "epoch": 1.7960795853233438,
      "grad_norm": 0.4991988241672516,
      "learning_rate": 8.194099818201184e-09,
      "loss": 0.136,
      "num_input_tokens_seen": 63816504,
      "step": 39675
    },
    {
      "epoch": 1.7963059372099868,
      "grad_norm": 0.3146803379058838,
      "learning_rate": 7.944687826400011e-09,
      "loss": 0.1401,
      "num_input_tokens_seen": 63824600,
      "step": 39680
    },
    {
      "epoch": 1.7965322890966298,
      "grad_norm": 0.5831406712532043,
      "learning_rate": 7.699129923599557e-09,
      "loss": 0.1783,
      "num_input_tokens_seen": 63833656,
      "step": 39685
    },
    {
      "epoch": 1.7967586409832728,
      "grad_norm": 0.8784933686256409,
      "learning_rate": 7.457426147663982e-09,
      "loss": 0.1191,
      "num_input_tokens_seen": 63841688,
      "step": 39690
    },
    {
      "epoch": 1.7969849928699155,
      "grad_norm": 0.4593794345855713,
      "learning_rate": 7.219576535871797e-09,
      "loss": 0.1621,
      "num_input_tokens_seen": 63849720,
      "step": 39695
    },
    {
      "epoch": 1.7972113447565585,
      "grad_norm": 0.5455235242843628,
      "learning_rate": 6.985581124896445e-09,
      "loss": 0.1218,
      "num_input_tokens_seen": 63857592,
      "step": 39700
    },
    {
      "epoch": 1.7974376966432015,
      "grad_norm": 0.9495696425437927,
      "learning_rate": 6.755439950828501e-09,
      "loss": 0.1426,
      "num_input_tokens_seen": 63865496,
      "step": 39705
    },
    {
      "epoch": 1.7976640485298445,
      "grad_norm": 0.2544742822647095,
      "learning_rate": 6.5291530491562444e-09,
      "loss": 0.1194,
      "num_input_tokens_seen": 63874104,
      "step": 39710
    },
    {
      "epoch": 1.7978904004164875,
      "grad_norm": 0.8620237708091736,
      "learning_rate": 6.3067204547739845e-09,
      "loss": 0.1499,
      "num_input_tokens_seen": 63882008,
      "step": 39715
    },
    {
      "epoch": 1.7981167523031303,
      "grad_norm": 0.7411177158355713,
      "learning_rate": 6.088142201987612e-09,
      "loss": 0.1694,
      "num_input_tokens_seen": 63890168,
      "step": 39720
    },
    {
      "epoch": 1.7983431041897733,
      "grad_norm": 0.7542753219604492,
      "learning_rate": 5.873418324503499e-09,
      "loss": 0.1266,
      "num_input_tokens_seen": 63898200,
      "step": 39725
    },
    {
      "epoch": 1.7985694560764163,
      "grad_norm": 0.336129754781723,
      "learning_rate": 5.6625488554340465e-09,
      "loss": 0.1216,
      "num_input_tokens_seen": 63906712,
      "step": 39730
    },
    {
      "epoch": 1.7987958079630593,
      "grad_norm": 0.49268755316734314,
      "learning_rate": 5.455533827297688e-09,
      "loss": 0.1081,
      "num_input_tokens_seen": 63914616,
      "step": 39735
    },
    {
      "epoch": 1.7990221598497023,
      "grad_norm": 0.456313818693161,
      "learning_rate": 5.252373272018885e-09,
      "loss": 0.1326,
      "num_input_tokens_seen": 63923096,
      "step": 39740
    },
    {
      "epoch": 1.7992485117363453,
      "grad_norm": 0.5540375113487244,
      "learning_rate": 5.053067220925356e-09,
      "loss": 0.1483,
      "num_input_tokens_seen": 63931352,
      "step": 39745
    },
    {
      "epoch": 1.7994748636229883,
      "grad_norm": 0.767997682094574,
      "learning_rate": 4.857615704759177e-09,
      "loss": 0.1827,
      "num_input_tokens_seen": 63938904,
      "step": 39750
    },
    {
      "epoch": 1.7997012155096312,
      "grad_norm": 0.383096307516098,
      "learning_rate": 4.666018753654577e-09,
      "loss": 0.1312,
      "num_input_tokens_seen": 63946968,
      "step": 39755
    },
    {
      "epoch": 1.7999275673962742,
      "grad_norm": 1.2012393474578857,
      "learning_rate": 4.478276397162917e-09,
      "loss": 0.1452,
      "num_input_tokens_seen": 63954424,
      "step": 39760
    },
    {
      "epoch": 1.8001539192829172,
      "grad_norm": 0.9878010749816895,
      "learning_rate": 4.294388664233262e-09,
      "loss": 0.1441,
      "num_input_tokens_seen": 63962552,
      "step": 39765
    },
    {
      "epoch": 1.8003802711695602,
      "grad_norm": 1.0522069931030273,
      "learning_rate": 4.114355583223484e-09,
      "loss": 0.178,
      "num_input_tokens_seen": 63970776,
      "step": 39770
    },
    {
      "epoch": 1.8006066230562032,
      "grad_norm": 0.7086911797523499,
      "learning_rate": 3.9381771818974845e-09,
      "loss": 0.1246,
      "num_input_tokens_seen": 63978616,
      "step": 39775
    },
    {
      "epoch": 1.8008329749428462,
      "grad_norm": 0.650505542755127,
      "learning_rate": 3.765853487427973e-09,
      "loss": 0.1518,
      "num_input_tokens_seen": 63986040,
      "step": 39780
    },
    {
      "epoch": 1.8010593268294892,
      "grad_norm": 0.7301549315452576,
      "learning_rate": 3.5973845263825857e-09,
      "loss": 0.1582,
      "num_input_tokens_seen": 63993816,
      "step": 39785
    },
    {
      "epoch": 1.8012856787161322,
      "grad_norm": 0.2986142039299011,
      "learning_rate": 3.4327703247488684e-09,
      "loss": 0.1029,
      "num_input_tokens_seen": 64001432,
      "step": 39790
    },
    {
      "epoch": 1.8015120306027752,
      "grad_norm": 0.458973228931427,
      "learning_rate": 3.2720109079037443e-09,
      "loss": 0.1658,
      "num_input_tokens_seen": 64009016,
      "step": 39795
    },
    {
      "epoch": 1.8017383824894182,
      "grad_norm": 0.44155973196029663,
      "learning_rate": 3.1151063006468193e-09,
      "loss": 0.1161,
      "num_input_tokens_seen": 64016568,
      "step": 39800
    },
    {
      "epoch": 1.8017383824894182,
      "eval_loss": 0.14332252740859985,
      "eval_runtime": 405.5482,
      "eval_samples_per_second": 96.834,
      "eval_steps_per_second": 24.209,
      "num_input_tokens_seen": 64016568,
      "step": 39800
    },
    {
      "epoch": 1.801964734376061,
      "grad_norm": 1.0966001749038696,
      "learning_rate": 2.962056527169854e-09,
      "loss": 0.1388,
      "num_input_tokens_seen": 64024664,
      "step": 39805
    },
    {
      "epoch": 1.802191086262704,
      "grad_norm": 1.0548237562179565,
      "learning_rate": 2.8128616110761898e-09,
      "loss": 0.1912,
      "num_input_tokens_seen": 64033176,
      "step": 39810
    },
    {
      "epoch": 1.802417438149347,
      "grad_norm": 0.8429155349731445,
      "learning_rate": 2.6675215753724223e-09,
      "loss": 0.1531,
      "num_input_tokens_seen": 64040888,
      "step": 39815
    },
    {
      "epoch": 1.80264379003599,
      "grad_norm": 0.6553663611412048,
      "learning_rate": 2.5260364424739557e-09,
      "loss": 0.1348,
      "num_input_tokens_seen": 64048504,
      "step": 39820
    },
    {
      "epoch": 1.802870141922633,
      "grad_norm": 0.5716578960418701,
      "learning_rate": 2.3884062341994475e-09,
      "loss": 0.1228,
      "num_input_tokens_seen": 64056088,
      "step": 39825
    },
    {
      "epoch": 1.803096493809276,
      "grad_norm": 0.5671219229698181,
      "learning_rate": 2.25463097177081e-09,
      "loss": 0.1471,
      "num_input_tokens_seen": 64064216,
      "step": 39830
    },
    {
      "epoch": 1.8033228456959187,
      "grad_norm": 0.5542780160903931,
      "learning_rate": 2.1247106758215397e-09,
      "loss": 0.1048,
      "num_input_tokens_seen": 64072280,
      "step": 39835
    },
    {
      "epoch": 1.8035491975825617,
      "grad_norm": 0.7130243182182312,
      "learning_rate": 1.998645366382834e-09,
      "loss": 0.1643,
      "num_input_tokens_seen": 64080600,
      "step": 39840
    },
    {
      "epoch": 1.8037755494692047,
      "grad_norm": 0.7182740569114685,
      "learning_rate": 1.876435062897475e-09,
      "loss": 0.137,
      "num_input_tokens_seen": 64088728,
      "step": 39845
    },
    {
      "epoch": 1.8040019013558477,
      "grad_norm": 0.32185098528862,
      "learning_rate": 1.758079784211497e-09,
      "loss": 0.1274,
      "num_input_tokens_seen": 64096504,
      "step": 39850
    },
    {
      "epoch": 1.8042282532424907,
      "grad_norm": 0.5083394646644592,
      "learning_rate": 1.6435795485797434e-09,
      "loss": 0.1226,
      "num_input_tokens_seen": 64104792,
      "step": 39855
    },
    {
      "epoch": 1.8044546051291337,
      "grad_norm": 0.4657995104789734,
      "learning_rate": 1.5329343736547596e-09,
      "loss": 0.1413,
      "num_input_tokens_seen": 64112600,
      "step": 39860
    },
    {
      "epoch": 1.8046809570157767,
      "grad_norm": 1.1466784477233887,
      "learning_rate": 1.4261442765006739e-09,
      "loss": 0.1529,
      "num_input_tokens_seen": 64120856,
      "step": 39865
    },
    {
      "epoch": 1.8049073089024197,
      "grad_norm": 0.6573595404624939,
      "learning_rate": 1.3232092735876445e-09,
      "loss": 0.1913,
      "num_input_tokens_seen": 64128824,
      "step": 39870
    },
    {
      "epoch": 1.8051336607890627,
      "grad_norm": 0.34706780314445496,
      "learning_rate": 1.2241293807918607e-09,
      "loss": 0.099,
      "num_input_tokens_seen": 64136664,
      "step": 39875
    },
    {
      "epoch": 1.8053600126757057,
      "grad_norm": 0.662437915802002,
      "learning_rate": 1.128904613387216e-09,
      "loss": 0.1757,
      "num_input_tokens_seen": 64144248,
      "step": 39880
    },
    {
      "epoch": 1.8055863645623487,
      "grad_norm": 0.6543940305709839,
      "learning_rate": 1.0375349860591853e-09,
      "loss": 0.1521,
      "num_input_tokens_seen": 64152600,
      "step": 39885
    },
    {
      "epoch": 1.8058127164489917,
      "grad_norm": 0.7714243531227112,
      "learning_rate": 9.5002051290205e-10,
      "loss": 0.1331,
      "num_input_tokens_seen": 64160440,
      "step": 39890
    },
    {
      "epoch": 1.8060390683356347,
      "grad_norm": 0.5656288862228394,
      "learning_rate": 8.663612074077954e-10,
      "loss": 0.1227,
      "num_input_tokens_seen": 64168472,
      "step": 39895
    },
    {
      "epoch": 1.8062654202222777,
      "grad_norm": 0.9270511269569397,
      "learning_rate": 7.865570824799884e-10,
      "loss": 0.1508,
      "num_input_tokens_seen": 64177208,
      "step": 39900
    },
    {
      "epoch": 1.8064917721089206,
      "grad_norm": 0.4457412660121918,
      "learning_rate": 7.106081504254514e-10,
      "loss": 0.133,
      "num_input_tokens_seen": 64185496,
      "step": 39905
    },
    {
      "epoch": 1.8067181239955636,
      "grad_norm": 0.44697558879852295,
      "learning_rate": 6.385144229570372e-10,
      "loss": 0.1528,
      "num_input_tokens_seen": 64193784,
      "step": 39910
    },
    {
      "epoch": 1.8069444758822066,
      "grad_norm": 0.7554659247398376,
      "learning_rate": 5.70275911190854e-10,
      "loss": 0.156,
      "num_input_tokens_seen": 64201208,
      "step": 39915
    },
    {
      "epoch": 1.8071708277688494,
      "grad_norm": 0.4540250599384308,
      "learning_rate": 5.058926256490403e-10,
      "loss": 0.1334,
      "num_input_tokens_seen": 64210104,
      "step": 39920
    },
    {
      "epoch": 1.8073971796554924,
      "grad_norm": 0.5049697160720825,
      "learning_rate": 4.4536457626254134e-10,
      "loss": 0.1461,
      "num_input_tokens_seen": 64218456,
      "step": 39925
    },
    {
      "epoch": 1.8076235315421354,
      "grad_norm": 0.7315605878829956,
      "learning_rate": 3.88691772365557e-10,
      "loss": 0.1207,
      "num_input_tokens_seen": 64226488,
      "step": 39930
    },
    {
      "epoch": 1.8078498834287784,
      "grad_norm": 0.5763887166976929,
      "learning_rate": 3.358742226955425e-10,
      "loss": 0.1092,
      "num_input_tokens_seen": 64234040,
      "step": 39935
    },
    {
      "epoch": 1.8080762353154214,
      "grad_norm": 0.7424412965774536,
      "learning_rate": 2.8691193539875925e-10,
      "loss": 0.1412,
      "num_input_tokens_seen": 64241912,
      "step": 39940
    },
    {
      "epoch": 1.8083025872020644,
      "grad_norm": 0.8532562851905823,
      "learning_rate": 2.418049180274995e-10,
      "loss": 0.1709,
      "num_input_tokens_seen": 64249912,
      "step": 39945
    },
    {
      "epoch": 1.8085289390887072,
      "grad_norm": 0.35048070549964905,
      "learning_rate": 2.005531775373104e-10,
      "loss": 0.158,
      "num_input_tokens_seen": 64257752,
      "step": 39950
    },
    {
      "epoch": 1.8087552909753501,
      "grad_norm": 0.5866716504096985,
      "learning_rate": 1.6315672028699435e-10,
      "loss": 0.1184,
      "num_input_tokens_seen": 64265272,
      "step": 39955
    },
    {
      "epoch": 1.8089816428619931,
      "grad_norm": 0.6328115463256836,
      "learning_rate": 1.2961555204693555e-10,
      "loss": 0.123,
      "num_input_tokens_seen": 64273304,
      "step": 39960
    },
    {
      "epoch": 1.8092079947486361,
      "grad_norm": 0.7203518748283386,
      "learning_rate": 9.992967798799768e-11,
      "loss": 0.1418,
      "num_input_tokens_seen": 64281784,
      "step": 39965
    },
    {
      "epoch": 1.8094343466352791,
      "grad_norm": 0.8972119092941284,
      "learning_rate": 7.409910268707521e-11,
      "loss": 0.14,
      "num_input_tokens_seen": 64289496,
      "step": 39970
    },
    {
      "epoch": 1.8096606985219221,
      "grad_norm": 0.5732694864273071,
      "learning_rate": 5.212383012986877e-11,
      "loss": 0.1756,
      "num_input_tokens_seen": 64297816,
      "step": 39975
    },
    {
      "epoch": 1.8098870504085651,
      "grad_norm": 0.7047970294952393,
      "learning_rate": 3.400386370533415e-11,
      "loss": 0.1588,
      "num_input_tokens_seen": 64306264,
      "step": 39980
    },
    {
      "epoch": 1.8101134022952081,
      "grad_norm": 0.8849195837974548,
      "learning_rate": 1.9739206205682258e-11,
      "loss": 0.1168,
      "num_input_tokens_seen": 64314168,
      "step": 39985
    },
    {
      "epoch": 1.8103397541818511,
      "grad_norm": 0.422608882188797,
      "learning_rate": 9.329859829154685e-12,
      "loss": 0.1419,
      "num_input_tokens_seen": 64321784,
      "step": 39990
    },
    {
      "epoch": 1.810566106068494,
      "grad_norm": 0.6595233678817749,
      "learning_rate": 2.7758261855748148e-12,
      "loss": 0.1172,
      "num_input_tokens_seen": 64330072,
      "step": 39995
    },
    {
      "epoch": 1.810792457955137,
      "grad_norm": 0.328424334526062,
      "learning_rate": 7.710628524559838e-14,
      "loss": 0.1234,
      "num_input_tokens_seen": 64338040,
      "step": 40000
    },
    {
      "epoch": 1.810792457955137,
      "eval_loss": 0.14330105483531952,
      "eval_runtime": 404.6772,
      "eval_samples_per_second": 97.043,
      "eval_steps_per_second": 24.261,
      "num_input_tokens_seen": 64338040,
      "step": 40000
    },
    {
      "epoch": 1.810792457955137,
      "num_input_tokens_seen": 64338040,
      "step": 40000,
      "total_flos": 2.6945428326475776e+17,
      "train_loss": 0.17023187619000674,
      "train_runtime": 106459.0209,
      "train_samples_per_second": 6.012,
      "train_steps_per_second": 0.376
    }
  ],
  "logging_steps": 5,
  "max_steps": 40000,
  "num_input_tokens_seen": 64338040,
  "num_train_epochs": 2,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6945428326475776e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}