| { | |
| "best_metric": 0.7897204756736755, | |
| "best_model_checkpoint": "./model_fine-tune/glot/mbert/ron-Latn/checkpoint-98000", | |
| "epoch": 13.435700575815739, | |
| "eval_steps": 500, | |
| "global_step": 98000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06854949273375377, | |
| "grad_norm": 4.000861644744873, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.7568, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06854949273375377, | |
| "eval_accuracy": 0.7095725691837217, | |
| "eval_loss": 1.6891406774520874, | |
| "eval_runtime": 259.6177, | |
| "eval_samples_per_second": 156.391, | |
| "eval_steps_per_second": 4.888, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13709898546750754, | |
| "grad_norm": 4.238007545471191, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.5655, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13709898546750754, | |
| "eval_accuracy": 0.7265132414723929, | |
| "eval_loss": 1.5632870197296143, | |
| "eval_runtime": 260.4774, | |
| "eval_samples_per_second": 155.875, | |
| "eval_steps_per_second": 4.872, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2056484782012613, | |
| "grad_norm": 3.5532774925231934, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.4944, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2056484782012613, | |
| "eval_accuracy": 0.7360105726108715, | |
| "eval_loss": 1.4989327192306519, | |
| "eval_runtime": 260.0937, | |
| "eval_samples_per_second": 156.105, | |
| "eval_steps_per_second": 4.879, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2741979709350151, | |
| "grad_norm": 3.5770881175994873, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.4324, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.2741979709350151, | |
| "eval_accuracy": 0.743094448707599, | |
| "eval_loss": 1.4517085552215576, | |
| "eval_runtime": 260.1265, | |
| "eval_samples_per_second": 156.086, | |
| "eval_steps_per_second": 4.878, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34274746366876885, | |
| "grad_norm": 3.5476059913635254, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.3797, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.34274746366876885, | |
| "eval_accuracy": 0.7503641171887161, | |
| "eval_loss": 1.4144247770309448, | |
| "eval_runtime": 258.9987, | |
| "eval_samples_per_second": 156.765, | |
| "eval_steps_per_second": 4.9, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4112969564025226, | |
| "grad_norm": 3.64892578125, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.3435, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4112969564025226, | |
| "eval_accuracy": 0.7549876281769841, | |
| "eval_loss": 1.380112648010254, | |
| "eval_runtime": 258.9885, | |
| "eval_samples_per_second": 156.771, | |
| "eval_steps_per_second": 4.9, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4798464491362764, | |
| "grad_norm": 3.1375253200531006, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.3133, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.4798464491362764, | |
| "eval_accuracy": 0.7577493236618825, | |
| "eval_loss": 1.374040126800537, | |
| "eval_runtime": 259.0671, | |
| "eval_samples_per_second": 156.724, | |
| "eval_steps_per_second": 4.898, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5483959418700302, | |
| "grad_norm": 3.1487910747528076, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2891, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5483959418700302, | |
| "eval_accuracy": 0.76247168256945, | |
| "eval_loss": 1.34718918800354, | |
| "eval_runtime": 259.0025, | |
| "eval_samples_per_second": 156.763, | |
| "eval_steps_per_second": 4.9, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6169454346037839, | |
| "grad_norm": 2.947143316268921, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.2764, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6169454346037839, | |
| "eval_accuracy": 0.7652144022180228, | |
| "eval_loss": 1.3174166679382324, | |
| "eval_runtime": 258.6684, | |
| "eval_samples_per_second": 156.965, | |
| "eval_steps_per_second": 4.906, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.6854949273375377, | |
| "grad_norm": 3.274010181427002, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.2492, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.6854949273375377, | |
| "eval_accuracy": 0.7671819928468485, | |
| "eval_loss": 1.313983678817749, | |
| "eval_runtime": 258.6625, | |
| "eval_samples_per_second": 156.969, | |
| "eval_steps_per_second": 4.906, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7540444200712915, | |
| "grad_norm": 3.109161138534546, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.2329, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7540444200712915, | |
| "eval_accuracy": 0.7713562268262496, | |
| "eval_loss": 1.2868432998657227, | |
| "eval_runtime": 260.0165, | |
| "eval_samples_per_second": 156.152, | |
| "eval_steps_per_second": 4.88, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.8225939128050452, | |
| "grad_norm": 7.044505596160889, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.2232, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.8225939128050452, | |
| "eval_accuracy": 0.7731761649539602, | |
| "eval_loss": 1.2668291330337524, | |
| "eval_runtime": 258.9434, | |
| "eval_samples_per_second": 156.799, | |
| "eval_steps_per_second": 4.901, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.891143405538799, | |
| "grad_norm": 3.029754400253296, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.2117, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.891143405538799, | |
| "eval_accuracy": 0.7750832812825318, | |
| "eval_loss": 1.2626760005950928, | |
| "eval_runtime": 260.4421, | |
| "eval_samples_per_second": 155.896, | |
| "eval_steps_per_second": 4.872, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.9596928982725528, | |
| "grad_norm": 2.961531639099121, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 1.1924, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.9596928982725528, | |
| "eval_accuracy": 0.7771516510947648, | |
| "eval_loss": 1.2538079023361206, | |
| "eval_runtime": 260.0863, | |
| "eval_samples_per_second": 156.11, | |
| "eval_steps_per_second": 4.879, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.0282423910063065, | |
| "grad_norm": 3.120314121246338, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 1.1807, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0282423910063065, | |
| "eval_accuracy": 0.7786684520598107, | |
| "eval_loss": 1.247827410697937, | |
| "eval_runtime": 258.2089, | |
| "eval_samples_per_second": 157.245, | |
| "eval_steps_per_second": 4.915, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.0967918837400603, | |
| "grad_norm": 3.0152571201324463, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.1666, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0967918837400603, | |
| "eval_accuracy": 0.7806228052923374, | |
| "eval_loss": 1.2236727476119995, | |
| "eval_runtime": 258.1981, | |
| "eval_samples_per_second": 157.251, | |
| "eval_steps_per_second": 4.915, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.165341376473814, | |
| "grad_norm": 4.560582637786865, | |
| "learning_rate": 9.15e-05, | |
| "loss": 1.1582, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.165341376473814, | |
| "eval_accuracy": 0.7811331658792164, | |
| "eval_loss": 1.2169686555862427, | |
| "eval_runtime": 257.413, | |
| "eval_samples_per_second": 157.731, | |
| "eval_steps_per_second": 4.93, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.2338908692075679, | |
| "grad_norm": 2.940659523010254, | |
| "learning_rate": 9.1e-05, | |
| "loss": 1.1376, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.2338908692075679, | |
| "eval_accuracy": 0.7844609998371601, | |
| "eval_loss": 1.1992230415344238, | |
| "eval_runtime": 257.3928, | |
| "eval_samples_per_second": 157.743, | |
| "eval_steps_per_second": 4.93, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.3024403619413216, | |
| "grad_norm": 3.4134812355041504, | |
| "learning_rate": 9.05e-05, | |
| "loss": 1.1358, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.3024403619413216, | |
| "eval_accuracy": 0.7851634886822189, | |
| "eval_loss": 1.2106844186782837, | |
| "eval_runtime": 257.3145, | |
| "eval_samples_per_second": 157.791, | |
| "eval_steps_per_second": 4.932, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.3709898546750754, | |
| "grad_norm": 2.8850438594818115, | |
| "learning_rate": 9e-05, | |
| "loss": 1.1281, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.3709898546750754, | |
| "eval_accuracy": 0.7872160784699163, | |
| "eval_loss": 1.2045215368270874, | |
| "eval_runtime": 257.4649, | |
| "eval_samples_per_second": 157.699, | |
| "eval_steps_per_second": 4.929, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.4395393474088292, | |
| "grad_norm": 2.752389669418335, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 1.119, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.4395393474088292, | |
| "eval_accuracy": 0.7882877785798362, | |
| "eval_loss": 1.1866884231567383, | |
| "eval_runtime": 257.0898, | |
| "eval_samples_per_second": 157.929, | |
| "eval_steps_per_second": 4.936, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.508088840142583, | |
| "grad_norm": 3.4821131229400635, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 1.1052, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.508088840142583, | |
| "eval_accuracy": 0.7891213465142315, | |
| "eval_loss": 1.177182912826538, | |
| "eval_runtime": 257.3303, | |
| "eval_samples_per_second": 157.782, | |
| "eval_steps_per_second": 4.931, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.5766383328763367, | |
| "grad_norm": 2.641080379486084, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 1.0969, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.5766383328763367, | |
| "eval_accuracy": 0.790978984380183, | |
| "eval_loss": 1.1677839756011963, | |
| "eval_runtime": 257.8775, | |
| "eval_samples_per_second": 157.447, | |
| "eval_steps_per_second": 4.921, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.6451878256100905, | |
| "grad_norm": 2.670703172683716, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.0973, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.6451878256100905, | |
| "eval_accuracy": 0.7914186279008998, | |
| "eval_loss": 1.1657401323318481, | |
| "eval_runtime": 257.6044, | |
| "eval_samples_per_second": 157.614, | |
| "eval_steps_per_second": 4.926, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.7137373183438442, | |
| "grad_norm": 4.184514045715332, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.0931, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.7137373183438442, | |
| "eval_accuracy": 0.7924260144852202, | |
| "eval_loss": 1.1656056642532349, | |
| "eval_runtime": 257.7225, | |
| "eval_samples_per_second": 157.542, | |
| "eval_steps_per_second": 4.924, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.782286811077598, | |
| "grad_norm": 3.4420437812805176, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.0882, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.782286811077598, | |
| "eval_accuracy": 0.7937018737958299, | |
| "eval_loss": 1.159055471420288, | |
| "eval_runtime": 257.6253, | |
| "eval_samples_per_second": 157.601, | |
| "eval_steps_per_second": 4.926, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.8508363038113518, | |
| "grad_norm": 2.942854642868042, | |
| "learning_rate": 8.65e-05, | |
| "loss": 1.0802, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.8508363038113518, | |
| "eval_accuracy": 0.7948528675102821, | |
| "eval_loss": 1.1472080945968628, | |
| "eval_runtime": 257.8341, | |
| "eval_samples_per_second": 157.473, | |
| "eval_steps_per_second": 4.922, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.9193857965451055, | |
| "grad_norm": 2.382511854171753, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.0766, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.9193857965451055, | |
| "eval_accuracy": 0.7966625616136973, | |
| "eval_loss": 1.1392544507980347, | |
| "eval_runtime": 258.8105, | |
| "eval_samples_per_second": 156.879, | |
| "eval_steps_per_second": 4.903, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.9879352892788593, | |
| "grad_norm": 2.583773374557495, | |
| "learning_rate": 8.55e-05, | |
| "loss": 1.0719, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.9879352892788593, | |
| "eval_accuracy": 0.7972140563559621, | |
| "eval_loss": 1.1423135995864868, | |
| "eval_runtime": 257.6602, | |
| "eval_samples_per_second": 157.58, | |
| "eval_steps_per_second": 4.925, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 2.056484782012613, | |
| "grad_norm": 2.788512945175171, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.0553, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.056484782012613, | |
| "eval_accuracy": 0.7980805136441053, | |
| "eval_loss": 1.128947377204895, | |
| "eval_runtime": 257.7272, | |
| "eval_samples_per_second": 157.539, | |
| "eval_steps_per_second": 4.924, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 2.125034274746367, | |
| "grad_norm": 2.9311256408691406, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 1.0451, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.125034274746367, | |
| "eval_accuracy": 0.7988239883829299, | |
| "eval_loss": 1.1336219310760498, | |
| "eval_runtime": 258.1352, | |
| "eval_samples_per_second": 157.29, | |
| "eval_steps_per_second": 4.916, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 2.1935837674801206, | |
| "grad_norm": 3.1457791328430176, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.0423, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.1935837674801206, | |
| "eval_accuracy": 0.7996375561822584, | |
| "eval_loss": 1.1223351955413818, | |
| "eval_runtime": 258.998, | |
| "eval_samples_per_second": 156.766, | |
| "eval_steps_per_second": 4.9, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 2.2621332602138744, | |
| "grad_norm": 2.9234752655029297, | |
| "learning_rate": 8.35e-05, | |
| "loss": 1.0449, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.2621332602138744, | |
| "eval_accuracy": 0.8002099641553145, | |
| "eval_loss": 1.1050431728363037, | |
| "eval_runtime": 258.0627, | |
| "eval_samples_per_second": 157.334, | |
| "eval_steps_per_second": 4.917, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.330682752947628, | |
| "grad_norm": 2.9982569217681885, | |
| "learning_rate": 8.3e-05, | |
| "loss": 1.0312, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.330682752947628, | |
| "eval_accuracy": 0.8017049854116234, | |
| "eval_loss": 1.1168311834335327, | |
| "eval_runtime": 258.8729, | |
| "eval_samples_per_second": 156.841, | |
| "eval_steps_per_second": 4.902, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.399232245681382, | |
| "grad_norm": 2.63649845123291, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.0291, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.399232245681382, | |
| "eval_accuracy": 0.8025579020578063, | |
| "eval_loss": 1.1074473857879639, | |
| "eval_runtime": 257.7749, | |
| "eval_samples_per_second": 157.51, | |
| "eval_steps_per_second": 4.923, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.4677817384151357, | |
| "grad_norm": 3.122042417526245, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.0245, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.4677817384151357, | |
| "eval_accuracy": 0.802570536529715, | |
| "eval_loss": 1.1076058149337769, | |
| "eval_runtime": 258.567, | |
| "eval_samples_per_second": 157.027, | |
| "eval_steps_per_second": 4.908, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.5363312311488895, | |
| "grad_norm": 3.2931995391845703, | |
| "learning_rate": 8.15e-05, | |
| "loss": 1.0236, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.5363312311488895, | |
| "eval_accuracy": 0.803701212821904, | |
| "eval_loss": 1.092323660850525, | |
| "eval_runtime": 257.5769, | |
| "eval_samples_per_second": 157.631, | |
| "eval_steps_per_second": 4.927, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.6048807238826432, | |
| "grad_norm": 3.31691837310791, | |
| "learning_rate": 8.1e-05, | |
| "loss": 1.0218, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.6048807238826432, | |
| "eval_accuracy": 0.8040252089042342, | |
| "eval_loss": 1.0852642059326172, | |
| "eval_runtime": 258.7184, | |
| "eval_samples_per_second": 156.935, | |
| "eval_steps_per_second": 4.905, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.673430216616397, | |
| "grad_norm": 2.602132558822632, | |
| "learning_rate": 8.05e-05, | |
| "loss": 1.0101, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.673430216616397, | |
| "eval_accuracy": 0.8049598014763766, | |
| "eval_loss": 1.0851576328277588, | |
| "eval_runtime": 259.1257, | |
| "eval_samples_per_second": 156.688, | |
| "eval_steps_per_second": 4.897, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.741979709350151, | |
| "grad_norm": 2.6089420318603516, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0154, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.741979709350151, | |
| "eval_accuracy": 0.8060563142838197, | |
| "eval_loss": 1.0850160121917725, | |
| "eval_runtime": 257.7528, | |
| "eval_samples_per_second": 157.523, | |
| "eval_steps_per_second": 4.923, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.8105292020839046, | |
| "grad_norm": 2.57804536819458, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 1.0092, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.8105292020839046, | |
| "eval_accuracy": 0.8063966673859484, | |
| "eval_loss": 1.064876675605774, | |
| "eval_runtime": 257.3862, | |
| "eval_samples_per_second": 157.747, | |
| "eval_steps_per_second": 4.93, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.8790786948176583, | |
| "grad_norm": 2.919243097305298, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 0.9962, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.8790786948176583, | |
| "eval_accuracy": 0.8074074812889718, | |
| "eval_loss": 1.0758228302001953, | |
| "eval_runtime": 257.4397, | |
| "eval_samples_per_second": 157.715, | |
| "eval_steps_per_second": 4.929, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.947628187551412, | |
| "grad_norm": 2.7142975330352783, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 0.9949, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.947628187551412, | |
| "eval_accuracy": 0.8072095077707074, | |
| "eval_loss": 1.0723259449005127, | |
| "eval_runtime": 257.4852, | |
| "eval_samples_per_second": 157.687, | |
| "eval_steps_per_second": 4.928, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 3.016177680285166, | |
| "grad_norm": 2.461714267730713, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.9933, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.016177680285166, | |
| "eval_accuracy": 0.8092224160930719, | |
| "eval_loss": 1.0564687252044678, | |
| "eval_runtime": 262.5555, | |
| "eval_samples_per_second": 154.642, | |
| "eval_steps_per_second": 4.833, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 3.0847271730189196, | |
| "grad_norm": 3.128793716430664, | |
| "learning_rate": 7.75e-05, | |
| "loss": 0.9751, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.0847271730189196, | |
| "eval_accuracy": 0.8092534032416199, | |
| "eval_loss": 1.0642082691192627, | |
| "eval_runtime": 263.2784, | |
| "eval_samples_per_second": 154.217, | |
| "eval_steps_per_second": 4.82, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 3.1532766657526734, | |
| "grad_norm": 2.560393810272217, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.9792, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.1532766657526734, | |
| "eval_accuracy": 0.8098133712563933, | |
| "eval_loss": 1.0663542747497559, | |
| "eval_runtime": 279.6936, | |
| "eval_samples_per_second": 145.166, | |
| "eval_steps_per_second": 4.537, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 3.221826158486427, | |
| "grad_norm": 2.993088483810425, | |
| "learning_rate": 7.65e-05, | |
| "loss": 0.9671, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.221826158486427, | |
| "eval_accuracy": 0.8106374773041439, | |
| "eval_loss": 1.0509235858917236, | |
| "eval_runtime": 302.0862, | |
| "eval_samples_per_second": 134.405, | |
| "eval_steps_per_second": 4.201, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 3.290375651220181, | |
| "grad_norm": 2.796325445175171, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.9667, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.290375651220181, | |
| "eval_accuracy": 0.8117766899143959, | |
| "eval_loss": 1.0434768199920654, | |
| "eval_runtime": 302.1312, | |
| "eval_samples_per_second": 134.385, | |
| "eval_steps_per_second": 4.2, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 3.3589251439539347, | |
| "grad_norm": 3.067168712615967, | |
| "learning_rate": 7.55e-05, | |
| "loss": 0.9676, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.3589251439539347, | |
| "eval_accuracy": 0.811503557227325, | |
| "eval_loss": 1.0434749126434326, | |
| "eval_runtime": 303.0455, | |
| "eval_samples_per_second": 133.98, | |
| "eval_steps_per_second": 4.187, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 3.4274746366876885, | |
| "grad_norm": 6.064915657043457, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.9659, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.4274746366876885, | |
| "eval_accuracy": 0.8128527567965343, | |
| "eval_loss": 1.0416052341461182, | |
| "eval_runtime": 301.3987, | |
| "eval_samples_per_second": 134.712, | |
| "eval_steps_per_second": 4.21, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.4960241294214423, | |
| "grad_norm": 2.412940740585327, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 0.9534, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.4960241294214423, | |
| "eval_accuracy": 0.8132953171379946, | |
| "eval_loss": 1.0385822057724, | |
| "eval_runtime": 300.1799, | |
| "eval_samples_per_second": 135.259, | |
| "eval_steps_per_second": 4.227, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.564573622155196, | |
| "grad_norm": 2.9293601512908936, | |
| "learning_rate": 7.4e-05, | |
| "loss": 0.9579, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.564573622155196, | |
| "eval_accuracy": 0.8135014968664439, | |
| "eval_loss": 1.0341033935546875, | |
| "eval_runtime": 302.1905, | |
| "eval_samples_per_second": 134.359, | |
| "eval_steps_per_second": 4.199, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 3.63312311488895, | |
| "grad_norm": 2.3857290744781494, | |
| "learning_rate": 7.35e-05, | |
| "loss": 0.9562, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.63312311488895, | |
| "eval_accuracy": 0.8136511493390661, | |
| "eval_loss": 1.0400645732879639, | |
| "eval_runtime": 300.9192, | |
| "eval_samples_per_second": 134.927, | |
| "eval_steps_per_second": 4.217, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 3.7016726076227036, | |
| "grad_norm": 2.8844683170318604, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.9581, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.7016726076227036, | |
| "eval_accuracy": 0.814749558109949, | |
| "eval_loss": 1.0379000902175903, | |
| "eval_runtime": 300.8792, | |
| "eval_samples_per_second": 134.945, | |
| "eval_steps_per_second": 4.218, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 3.7702221003564573, | |
| "grad_norm": 3.2288286685943604, | |
| "learning_rate": 7.25e-05, | |
| "loss": 0.9524, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.7702221003564573, | |
| "eval_accuracy": 0.8149326212786671, | |
| "eval_loss": 1.0268869400024414, | |
| "eval_runtime": 303.3311, | |
| "eval_samples_per_second": 133.854, | |
| "eval_steps_per_second": 4.184, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 3.838771593090211, | |
| "grad_norm": 2.84405255317688, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.9366, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.838771593090211, | |
| "eval_accuracy": 0.8165028910386263, | |
| "eval_loss": 1.0258753299713135, | |
| "eval_runtime": 301.0226, | |
| "eval_samples_per_second": 134.88, | |
| "eval_steps_per_second": 4.216, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 3.907321085823965, | |
| "grad_norm": 2.78871488571167, | |
| "learning_rate": 7.15e-05, | |
| "loss": 0.9489, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.907321085823965, | |
| "eval_accuracy": 0.8153732124964858, | |
| "eval_loss": 1.0232901573181152, | |
| "eval_runtime": 303.0362, | |
| "eval_samples_per_second": 133.984, | |
| "eval_steps_per_second": 4.188, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.9758705785577186, | |
| "grad_norm": 2.6128122806549072, | |
| "learning_rate": 7.1e-05, | |
| "loss": 0.9372, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.9758705785577186, | |
| "eval_accuracy": 0.8167775740472835, | |
| "eval_loss": 1.0158660411834717, | |
| "eval_runtime": 305.0716, | |
| "eval_samples_per_second": 133.09, | |
| "eval_steps_per_second": 4.16, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 4.044420071291473, | |
| "grad_norm": 2.4649457931518555, | |
| "learning_rate": 7.05e-05, | |
| "loss": 0.9389, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.044420071291473, | |
| "eval_accuracy": 0.8169951094301297, | |
| "eval_loss": 1.0179492235183716, | |
| "eval_runtime": 302.9032, | |
| "eval_samples_per_second": 134.043, | |
| "eval_steps_per_second": 4.189, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 4.112969564025226, | |
| "grad_norm": 2.637385845184326, | |
| "learning_rate": 7e-05, | |
| "loss": 0.9224, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.112969564025226, | |
| "eval_accuracy": 0.8178961830392573, | |
| "eval_loss": 1.0067319869995117, | |
| "eval_runtime": 300.9419, | |
| "eval_samples_per_second": 134.916, | |
| "eval_steps_per_second": 4.217, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 4.18151905675898, | |
| "grad_norm": 2.863875389099121, | |
| "learning_rate": 6.95e-05, | |
| "loss": 0.9205, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.18151905675898, | |
| "eval_accuracy": 0.8184249147223241, | |
| "eval_loss": 1.0113714933395386, | |
| "eval_runtime": 301.5875, | |
| "eval_samples_per_second": 134.628, | |
| "eval_steps_per_second": 4.208, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 4.250068549492734, | |
| "grad_norm": 2.4322681427001953, | |
| "learning_rate": 6.9e-05, | |
| "loss": 0.9247, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.250068549492734, | |
| "eval_accuracy": 0.8186936629410247, | |
| "eval_loss": 1.003678798675537, | |
| "eval_runtime": 303.1297, | |
| "eval_samples_per_second": 133.943, | |
| "eval_steps_per_second": 4.186, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 4.318618042226488, | |
| "grad_norm": 2.998030424118042, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 0.9178, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.318618042226488, | |
| "eval_accuracy": 0.8182408058742606, | |
| "eval_loss": 1.0021617412567139, | |
| "eval_runtime": 302.4668, | |
| "eval_samples_per_second": 134.236, | |
| "eval_steps_per_second": 4.196, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 4.387167534960241, | |
| "grad_norm": 2.5294859409332275, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.9176, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.387167534960241, | |
| "eval_accuracy": 0.8199272207420885, | |
| "eval_loss": 1.0028120279312134, | |
| "eval_runtime": 301.3201, | |
| "eval_samples_per_second": 134.747, | |
| "eval_steps_per_second": 4.211, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 4.4557170276939955, | |
| "grad_norm": 4.368305206298828, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 0.9187, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.4557170276939955, | |
| "eval_accuracy": 0.8202180503701034, | |
| "eval_loss": 1.0021815299987793, | |
| "eval_runtime": 301.1897, | |
| "eval_samples_per_second": 134.805, | |
| "eval_steps_per_second": 4.213, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 4.524266520427749, | |
| "grad_norm": 3.3230433464050293, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.9169, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.524266520427749, | |
| "eval_accuracy": 0.8205231724765897, | |
| "eval_loss": 0.9979987740516663, | |
| "eval_runtime": 300.7001, | |
| "eval_samples_per_second": 135.025, | |
| "eval_steps_per_second": 4.22, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 4.592816013161503, | |
| "grad_norm": 2.592043876647949, | |
| "learning_rate": 6.65e-05, | |
| "loss": 0.9125, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 4.592816013161503, | |
| "eval_accuracy": 0.8206935126919565, | |
| "eval_loss": 0.9938598871231079, | |
| "eval_runtime": 301.9997, | |
| "eval_samples_per_second": 134.444, | |
| "eval_steps_per_second": 4.202, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 4.661365505895256, | |
| "grad_norm": 2.446427345275879, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.9146, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.661365505895256, | |
| "eval_accuracy": 0.8216566473447208, | |
| "eval_loss": 0.9849461913108826, | |
| "eval_runtime": 302.9739, | |
| "eval_samples_per_second": 134.012, | |
| "eval_steps_per_second": 4.188, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 4.72991499862901, | |
| "grad_norm": 2.884946346282959, | |
| "learning_rate": 6.55e-05, | |
| "loss": 0.9018, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.72991499862901, | |
| "eval_accuracy": 0.8217915724349476, | |
| "eval_loss": 1.0003894567489624, | |
| "eval_runtime": 300.4012, | |
| "eval_samples_per_second": 135.159, | |
| "eval_steps_per_second": 4.224, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 4.798464491362764, | |
| "grad_norm": 2.8886282444000244, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.9014, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.798464491362764, | |
| "eval_accuracy": 0.8224961720866916, | |
| "eval_loss": 0.9889456629753113, | |
| "eval_runtime": 301.0802, | |
| "eval_samples_per_second": 134.854, | |
| "eval_steps_per_second": 4.215, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 4.867013984096518, | |
| "grad_norm": 2.473068952560425, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 0.8919, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.867013984096518, | |
| "eval_accuracy": 0.822945545786959, | |
| "eval_loss": 0.9848706722259521, | |
| "eval_runtime": 303.9067, | |
| "eval_samples_per_second": 133.6, | |
| "eval_steps_per_second": 4.176, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 4.935563476830271, | |
| "grad_norm": 3.0716209411621094, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.8993, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 4.935563476830271, | |
| "eval_accuracy": 0.8222583681418539, | |
| "eval_loss": 0.9929753541946411, | |
| "eval_runtime": 299.5059, | |
| "eval_samples_per_second": 135.563, | |
| "eval_steps_per_second": 4.237, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 5.004112969564026, | |
| "grad_norm": 2.3323957920074463, | |
| "learning_rate": 6.35e-05, | |
| "loss": 0.9009, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 5.004112969564026, | |
| "eval_accuracy": 0.8241444257225273, | |
| "eval_loss": 0.9798668622970581, | |
| "eval_runtime": 299.3852, | |
| "eval_samples_per_second": 135.618, | |
| "eval_steps_per_second": 4.239, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 5.072662462297779, | |
| "grad_norm": 2.7152209281921387, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.8843, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 5.072662462297779, | |
| "eval_accuracy": 0.8234012575934279, | |
| "eval_loss": 0.9811968803405762, | |
| "eval_runtime": 300.6715, | |
| "eval_samples_per_second": 135.038, | |
| "eval_steps_per_second": 4.221, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 5.141211955031533, | |
| "grad_norm": 2.526486396789551, | |
| "learning_rate": 6.25e-05, | |
| "loss": 0.8846, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 5.141211955031533, | |
| "eval_accuracy": 0.8247555724795991, | |
| "eval_loss": 0.9730820655822754, | |
| "eval_runtime": 299.291, | |
| "eval_samples_per_second": 135.661, | |
| "eval_steps_per_second": 4.24, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 5.2097614477652865, | |
| "grad_norm": 2.5805063247680664, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.8807, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.2097614477652865, | |
| "eval_accuracy": 0.8250464067024924, | |
| "eval_loss": 0.9684708118438721, | |
| "eval_runtime": 302.4457, | |
| "eval_samples_per_second": 134.246, | |
| "eval_steps_per_second": 4.196, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 5.278310940499041, | |
| "grad_norm": 2.559605360031128, | |
| "learning_rate": 6.15e-05, | |
| "loss": 0.8802, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.278310940499041, | |
| "eval_accuracy": 0.8254996987535631, | |
| "eval_loss": 0.973818838596344, | |
| "eval_runtime": 300.146, | |
| "eval_samples_per_second": 135.274, | |
| "eval_steps_per_second": 4.228, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 5.346860433232794, | |
| "grad_norm": 2.1615304946899414, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.8789, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.346860433232794, | |
| "eval_accuracy": 0.8254549864960571, | |
| "eval_loss": 0.9578101634979248, | |
| "eval_runtime": 302.5653, | |
| "eval_samples_per_second": 134.193, | |
| "eval_steps_per_second": 4.194, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 5.415409925966548, | |
| "grad_norm": 2.2763609886169434, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.8843, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 5.415409925966548, | |
| "eval_accuracy": 0.8262787384248012, | |
| "eval_loss": 0.9698151350021362, | |
| "eval_runtime": 301.7649, | |
| "eval_samples_per_second": 134.548, | |
| "eval_steps_per_second": 4.205, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 5.483959418700302, | |
| "grad_norm": 2.3774330615997314, | |
| "learning_rate": 6e-05, | |
| "loss": 0.8714, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 5.483959418700302, | |
| "eval_accuracy": 0.8263406985859876, | |
| "eval_loss": 0.9681651592254639, | |
| "eval_runtime": 302.4298, | |
| "eval_samples_per_second": 134.253, | |
| "eval_steps_per_second": 4.196, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 5.552508911434055, | |
| "grad_norm": 2.3430614471435547, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.8676, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 5.552508911434055, | |
| "eval_accuracy": 0.8263813882499345, | |
| "eval_loss": 0.9515417814254761, | |
| "eval_runtime": 301.876, | |
| "eval_samples_per_second": 134.499, | |
| "eval_steps_per_second": 4.204, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 5.621058404167809, | |
| "grad_norm": 2.3059141635894775, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.8721, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 5.621058404167809, | |
| "eval_accuracy": 0.8275216377261674, | |
| "eval_loss": 0.962243914604187, | |
| "eval_runtime": 302.2447, | |
| "eval_samples_per_second": 134.335, | |
| "eval_steps_per_second": 4.199, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 5.689607896901563, | |
| "grad_norm": 2.462218999862671, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.8699, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 5.689607896901563, | |
| "eval_accuracy": 0.8283203266406056, | |
| "eval_loss": 0.9477165341377258, | |
| "eval_runtime": 302.7485, | |
| "eval_samples_per_second": 134.111, | |
| "eval_steps_per_second": 4.192, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 5.758157389635317, | |
| "grad_norm": 3.0347349643707275, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.8634, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.758157389635317, | |
| "eval_accuracy": 0.8281108707620414, | |
| "eval_loss": 0.9486715197563171, | |
| "eval_runtime": 304.225, | |
| "eval_samples_per_second": 133.46, | |
| "eval_steps_per_second": 4.171, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 5.82670688236907, | |
| "grad_norm": 3.0054922103881836, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.8743, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 5.82670688236907, | |
| "eval_accuracy": 0.8284311134181968, | |
| "eval_loss": 0.9539070725440979, | |
| "eval_runtime": 304.8553, | |
| "eval_samples_per_second": 133.185, | |
| "eval_steps_per_second": 4.163, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 5.895256375102824, | |
| "grad_norm": 2.29243540763855, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.8667, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.895256375102824, | |
| "eval_accuracy": 0.8291235685160401, | |
| "eval_loss": 0.9470139145851135, | |
| "eval_runtime": 298.7685, | |
| "eval_samples_per_second": 135.898, | |
| "eval_steps_per_second": 4.247, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 5.963805867836578, | |
| "grad_norm": 2.5743372440338135, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.8681, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 5.963805867836578, | |
| "eval_accuracy": 0.8291158725629887, | |
| "eval_loss": 0.946834921836853, | |
| "eval_runtime": 284.2736, | |
| "eval_samples_per_second": 142.827, | |
| "eval_steps_per_second": 4.464, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 6.032355360570332, | |
| "grad_norm": 2.33494234085083, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.8594, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 6.032355360570332, | |
| "eval_accuracy": 0.8301663716691428, | |
| "eval_loss": 0.9472524523735046, | |
| "eval_runtime": 261.9172, | |
| "eval_samples_per_second": 155.018, | |
| "eval_steps_per_second": 4.845, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 6.100904853304086, | |
| "grad_norm": 2.7616426944732666, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.8517, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 6.100904853304086, | |
| "eval_accuracy": 0.8304027916380742, | |
| "eval_loss": 0.9408496022224426, | |
| "eval_runtime": 257.341, | |
| "eval_samples_per_second": 157.775, | |
| "eval_steps_per_second": 4.931, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 6.169454346037839, | |
| "grad_norm": 2.6394338607788086, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.8453, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 6.169454346037839, | |
| "eval_accuracy": 0.8302381896975964, | |
| "eval_loss": 0.945652425289154, | |
| "eval_runtime": 257.4862, | |
| "eval_samples_per_second": 157.686, | |
| "eval_steps_per_second": 4.928, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 6.2380038387715935, | |
| "grad_norm": 2.6004316806793213, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.8486, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 6.2380038387715935, | |
| "eval_accuracy": 0.8311095745962099, | |
| "eval_loss": 0.940719485282898, | |
| "eval_runtime": 258.4002, | |
| "eval_samples_per_second": 157.128, | |
| "eval_steps_per_second": 4.911, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 6.306553331505347, | |
| "grad_norm": 2.722169876098633, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.8469, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 6.306553331505347, | |
| "eval_accuracy": 0.8303636681664364, | |
| "eval_loss": 0.9365447759628296, | |
| "eval_runtime": 257.5562, | |
| "eval_samples_per_second": 157.643, | |
| "eval_steps_per_second": 4.927, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 6.375102824239101, | |
| "grad_norm": 2.955397367477417, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.8434, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 6.375102824239101, | |
| "eval_accuracy": 0.8311330928241353, | |
| "eval_loss": 0.9382375478744507, | |
| "eval_runtime": 258.6675, | |
| "eval_samples_per_second": 156.966, | |
| "eval_steps_per_second": 4.906, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 6.443652316972854, | |
| "grad_norm": 2.375140428543091, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.8343, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 6.443652316972854, | |
| "eval_accuracy": 0.8315033650181133, | |
| "eval_loss": 0.934901773929596, | |
| "eval_runtime": 257.5199, | |
| "eval_samples_per_second": 157.665, | |
| "eval_steps_per_second": 4.928, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 6.512201809706609, | |
| "grad_norm": 2.4617624282836914, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.8312, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 6.512201809706609, | |
| "eval_accuracy": 0.8317886916287004, | |
| "eval_loss": 0.9246230721473694, | |
| "eval_runtime": 258.5693, | |
| "eval_samples_per_second": 157.026, | |
| "eval_steps_per_second": 4.908, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 6.580751302440362, | |
| "grad_norm": 2.4794909954071045, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.8365, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 6.580751302440362, | |
| "eval_accuracy": 0.8332747655954476, | |
| "eval_loss": 0.9223575592041016, | |
| "eval_runtime": 257.532, | |
| "eval_samples_per_second": 157.658, | |
| "eval_steps_per_second": 4.928, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 6.649300795174116, | |
| "grad_norm": 2.893775224685669, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.8307, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 6.649300795174116, | |
| "eval_accuracy": 0.8332412930945546, | |
| "eval_loss": 0.9224662184715271, | |
| "eval_runtime": 257.3988, | |
| "eval_samples_per_second": 157.74, | |
| "eval_steps_per_second": 4.93, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 6.717850287907869, | |
| "grad_norm": 2.4228713512420654, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.838, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 6.717850287907869, | |
| "eval_accuracy": 0.8337124591782815, | |
| "eval_loss": 0.9226129055023193, | |
| "eval_runtime": 258.4175, | |
| "eval_samples_per_second": 157.118, | |
| "eval_steps_per_second": 4.911, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 6.786399780641624, | |
| "grad_norm": 2.462571144104004, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.8355, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 6.786399780641624, | |
| "eval_accuracy": 0.8337023695286286, | |
| "eval_loss": 0.9333141446113586, | |
| "eval_runtime": 257.7508, | |
| "eval_samples_per_second": 157.524, | |
| "eval_steps_per_second": 4.923, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 6.854949273375377, | |
| "grad_norm": 2.5558559894561768, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8391, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 6.854949273375377, | |
| "eval_accuracy": 0.8339626635022332, | |
| "eval_loss": 0.9166584610939026, | |
| "eval_runtime": 258.4946, | |
| "eval_samples_per_second": 157.071, | |
| "eval_steps_per_second": 4.909, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 6.923498766109131, | |
| "grad_norm": 2.733778953552246, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.834, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 6.923498766109131, | |
| "eval_accuracy": 0.8344111438621534, | |
| "eval_loss": 0.9235773086547852, | |
| "eval_runtime": 258.6401, | |
| "eval_samples_per_second": 156.983, | |
| "eval_steps_per_second": 4.906, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 6.9920482588428845, | |
| "grad_norm": 2.802053451538086, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.8269, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 6.9920482588428845, | |
| "eval_accuracy": 0.8348735131396657, | |
| "eval_loss": 0.920405924320221, | |
| "eval_runtime": 257.5552, | |
| "eval_samples_per_second": 157.644, | |
| "eval_steps_per_second": 4.927, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 7.060597751576639, | |
| "grad_norm": 2.666555404663086, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.8085, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 7.060597751576639, | |
| "eval_accuracy": 0.8347147667297783, | |
| "eval_loss": 0.9154396057128906, | |
| "eval_runtime": 258.4259, | |
| "eval_samples_per_second": 157.113, | |
| "eval_steps_per_second": 4.91, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 7.129147244310392, | |
| "grad_norm": 2.4190189838409424, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.819, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 7.129147244310392, | |
| "eval_accuracy": 0.8344749377257509, | |
| "eval_loss": 0.9256834983825684, | |
| "eval_runtime": 258.4604, | |
| "eval_samples_per_second": 157.092, | |
| "eval_steps_per_second": 4.91, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 7.197696737044146, | |
| "grad_norm": 2.8802294731140137, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.8238, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 7.197696737044146, | |
| "eval_accuracy": 0.8351059340465287, | |
| "eval_loss": 0.9185708165168762, | |
| "eval_runtime": 258.4374, | |
| "eval_samples_per_second": 157.106, | |
| "eval_steps_per_second": 4.91, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 7.2662462297779, | |
| "grad_norm": 2.5864334106445312, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.8065, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 7.2662462297779, | |
| "eval_accuracy": 0.8358540639956455, | |
| "eval_loss": 0.9095313549041748, | |
| "eval_runtime": 258.6881, | |
| "eval_samples_per_second": 156.953, | |
| "eval_steps_per_second": 4.906, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 7.334795722511654, | |
| "grad_norm": 3.0811657905578613, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.8199, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 7.334795722511654, | |
| "eval_accuracy": 0.8364544665607716, | |
| "eval_loss": 0.9065914154052734, | |
| "eval_runtime": 258.6721, | |
| "eval_samples_per_second": 156.963, | |
| "eval_steps_per_second": 4.906, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 7.403345215245407, | |
| "grad_norm": 2.3102753162384033, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.8018, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 7.403345215245407, | |
| "eval_accuracy": 0.8359817384838574, | |
| "eval_loss": 0.9075337052345276, | |
| "eval_runtime": 257.5046, | |
| "eval_samples_per_second": 157.675, | |
| "eval_steps_per_second": 4.928, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 7.471894707979161, | |
| "grad_norm": 2.269843578338623, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.8102, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 7.471894707979161, | |
| "eval_accuracy": 0.8372314644838865, | |
| "eval_loss": 0.9001559019088745, | |
| "eval_runtime": 258.4508, | |
| "eval_samples_per_second": 157.098, | |
| "eval_steps_per_second": 4.91, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 7.540444200712915, | |
| "grad_norm": 2.617309093475342, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8194, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 7.540444200712915, | |
| "eval_accuracy": 0.8369146258637555, | |
| "eval_loss": 0.899241030216217, | |
| "eval_runtime": 257.4931, | |
| "eval_samples_per_second": 157.682, | |
| "eval_steps_per_second": 4.928, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 7.608993693446669, | |
| "grad_norm": 2.4634625911712646, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.8138, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 7.608993693446669, | |
| "eval_accuracy": 0.8369996392235467, | |
| "eval_loss": 0.9033562541007996, | |
| "eval_runtime": 258.54, | |
| "eval_samples_per_second": 157.043, | |
| "eval_steps_per_second": 4.908, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 7.677543186180422, | |
| "grad_norm": 2.850604772567749, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.8077, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 7.677543186180422, | |
| "eval_accuracy": 0.8378181332713783, | |
| "eval_loss": 0.9002473950386047, | |
| "eval_runtime": 257.5581, | |
| "eval_samples_per_second": 157.642, | |
| "eval_steps_per_second": 4.927, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 7.746092678914176, | |
| "grad_norm": 2.3677656650543213, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.8119, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 7.746092678914176, | |
| "eval_accuracy": 0.8382744553424182, | |
| "eval_loss": 0.8992937803268433, | |
| "eval_runtime": 258.5324, | |
| "eval_samples_per_second": 157.048, | |
| "eval_steps_per_second": 4.908, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 7.81464217164793, | |
| "grad_norm": 2.0961389541625977, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.8029, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 7.81464217164793, | |
| "eval_accuracy": 0.8377691746192005, | |
| "eval_loss": 0.8913019895553589, | |
| "eval_runtime": 258.5294, | |
| "eval_samples_per_second": 157.05, | |
| "eval_steps_per_second": 4.909, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 7.883191664381684, | |
| "grad_norm": 2.424496650695801, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.802, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 7.883191664381684, | |
| "eval_accuracy": 0.8387740254965514, | |
| "eval_loss": 0.8913179039955139, | |
| "eval_runtime": 258.6018, | |
| "eval_samples_per_second": 157.006, | |
| "eval_steps_per_second": 4.907, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 7.951741157115437, | |
| "grad_norm": 2.8273098468780518, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.7887, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 7.951741157115437, | |
| "eval_accuracy": 0.8389023572318363, | |
| "eval_loss": 0.8917869329452515, | |
| "eval_runtime": 259.2663, | |
| "eval_samples_per_second": 156.603, | |
| "eval_steps_per_second": 4.895, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 8.02029064984919, | |
| "grad_norm": 2.5863022804260254, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.7902, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 8.02029064984919, | |
| "eval_accuracy": 0.8385434985627828, | |
| "eval_loss": 0.8866747617721558, | |
| "eval_runtime": 257.9905, | |
| "eval_samples_per_second": 157.378, | |
| "eval_steps_per_second": 4.919, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 8.088840142582946, | |
| "grad_norm": 2.357172727584839, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.7892, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 8.088840142582946, | |
| "eval_accuracy": 0.8392591622264518, | |
| "eval_loss": 0.8825114369392395, | |
| "eval_runtime": 258.6746, | |
| "eval_samples_per_second": 156.962, | |
| "eval_steps_per_second": 4.906, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 8.157389635316699, | |
| "grad_norm": 2.3970017433166504, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.7928, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 8.157389635316699, | |
| "eval_accuracy": 0.8398791985985096, | |
| "eval_loss": 0.8858514428138733, | |
| "eval_runtime": 258.8038, | |
| "eval_samples_per_second": 156.883, | |
| "eval_steps_per_second": 4.903, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 8.225939128050452, | |
| "grad_norm": 2.1816744804382324, | |
| "learning_rate": 4e-05, | |
| "loss": 0.786, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 8.225939128050452, | |
| "eval_accuracy": 0.8399906549712216, | |
| "eval_loss": 0.8871041536331177, | |
| "eval_runtime": 257.814, | |
| "eval_samples_per_second": 157.486, | |
| "eval_steps_per_second": 4.922, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 8.294488620784206, | |
| "grad_norm": 2.6891512870788574, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.7838, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 8.294488620784206, | |
| "eval_accuracy": 0.8402320722657605, | |
| "eval_loss": 0.8783635497093201, | |
| "eval_runtime": 257.6277, | |
| "eval_samples_per_second": 157.6, | |
| "eval_steps_per_second": 4.926, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 8.36303811351796, | |
| "grad_norm": 2.2459070682525635, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.7857, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 8.36303811351796, | |
| "eval_accuracy": 0.8401854172212088, | |
| "eval_loss": 0.8782520294189453, | |
| "eval_runtime": 258.8529, | |
| "eval_samples_per_second": 156.854, | |
| "eval_steps_per_second": 4.902, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 8.431587606251714, | |
| "grad_norm": 2.3516621589660645, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.7807, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 8.431587606251714, | |
| "eval_accuracy": 0.8408464558399127, | |
| "eval_loss": 0.8879706263542175, | |
| "eval_runtime": 258.721, | |
| "eval_samples_per_second": 156.934, | |
| "eval_steps_per_second": 4.905, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 8.500137098985467, | |
| "grad_norm": 2.408498764038086, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.7869, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 8.500137098985467, | |
| "eval_accuracy": 0.8411305726347277, | |
| "eval_loss": 0.8754673004150391, | |
| "eval_runtime": 258.7894, | |
| "eval_samples_per_second": 156.892, | |
| "eval_steps_per_second": 4.904, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 8.56868659171922, | |
| "grad_norm": 2.7398715019226074, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.7768, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 8.56868659171922, | |
| "eval_accuracy": 0.8415116048695496, | |
| "eval_loss": 0.8766404390335083, | |
| "eval_runtime": 258.7705, | |
| "eval_samples_per_second": 156.903, | |
| "eval_steps_per_second": 4.904, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 8.637236084452976, | |
| "grad_norm": 2.7151975631713867, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.7806, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 8.637236084452976, | |
| "eval_accuracy": 0.8416668834829366, | |
| "eval_loss": 0.8836163282394409, | |
| "eval_runtime": 257.8541, | |
| "eval_samples_per_second": 157.461, | |
| "eval_steps_per_second": 4.921, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 8.70578557718673, | |
| "grad_norm": 2.50140380859375, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.7811, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 8.70578557718673, | |
| "eval_accuracy": 0.8423243126907809, | |
| "eval_loss": 0.8704027533531189, | |
| "eval_runtime": 256.7585, | |
| "eval_samples_per_second": 158.133, | |
| "eval_steps_per_second": 4.942, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 8.774335069920483, | |
| "grad_norm": 2.9045302867889404, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.7733, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 8.774335069920483, | |
| "eval_accuracy": 0.8424119193113302, | |
| "eval_loss": 0.8675287365913391, | |
| "eval_runtime": 258.0076, | |
| "eval_samples_per_second": 157.367, | |
| "eval_steps_per_second": 4.918, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 8.842884562654236, | |
| "grad_norm": 2.8266477584838867, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.7728, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 8.842884562654236, | |
| "eval_accuracy": 0.8425746458723948, | |
| "eval_loss": 0.8766723871231079, | |
| "eval_runtime": 257.1101, | |
| "eval_samples_per_second": 157.917, | |
| "eval_steps_per_second": 4.936, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 8.911434055387991, | |
| "grad_norm": 2.3678436279296875, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.7779, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 8.911434055387991, | |
| "eval_accuracy": 0.842695820921136, | |
| "eval_loss": 0.8741580843925476, | |
| "eval_runtime": 258.4977, | |
| "eval_samples_per_second": 157.069, | |
| "eval_steps_per_second": 4.909, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 8.979983548121744, | |
| "grad_norm": 2.4681596755981445, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.7779, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 8.979983548121744, | |
| "eval_accuracy": 0.8435885693668189, | |
| "eval_loss": 0.8590840697288513, | |
| "eval_runtime": 258.179, | |
| "eval_samples_per_second": 157.263, | |
| "eval_steps_per_second": 4.915, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 9.048533040855498, | |
| "grad_norm": 2.4200708866119385, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.7704, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 9.048533040855498, | |
| "eval_accuracy": 0.8437169036285643, | |
| "eval_loss": 0.8766728639602661, | |
| "eval_runtime": 258.0969, | |
| "eval_samples_per_second": 157.313, | |
| "eval_steps_per_second": 4.917, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 9.117082533589251, | |
| "grad_norm": 2.753324270248413, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.7695, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 9.117082533589251, | |
| "eval_accuracy": 0.8435838175840091, | |
| "eval_loss": 0.881564199924469, | |
| "eval_runtime": 259.2788, | |
| "eval_samples_per_second": 156.596, | |
| "eval_steps_per_second": 4.894, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 9.185632026323006, | |
| "grad_norm": 2.490852117538452, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.7617, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 9.185632026323006, | |
| "eval_accuracy": 0.8437435512372087, | |
| "eval_loss": 0.8751281499862671, | |
| "eval_runtime": 256.1107, | |
| "eval_samples_per_second": 158.533, | |
| "eval_steps_per_second": 4.955, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 9.25418151905676, | |
| "grad_norm": 2.581777334213257, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.7585, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 9.25418151905676, | |
| "eval_accuracy": 0.8439763030486503, | |
| "eval_loss": 0.8657551407814026, | |
| "eval_runtime": 257.7079, | |
| "eval_samples_per_second": 157.55, | |
| "eval_steps_per_second": 4.924, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 9.322731011790513, | |
| "grad_norm": 2.997283458709717, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.7657, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 9.322731011790513, | |
| "eval_accuracy": 0.8446350018812996, | |
| "eval_loss": 0.8639153838157654, | |
| "eval_runtime": 256.5945, | |
| "eval_samples_per_second": 158.234, | |
| "eval_steps_per_second": 4.946, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 9.391280504524266, | |
| "grad_norm": 2.7763564586639404, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.759, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 9.391280504524266, | |
| "eval_accuracy": 0.8448020556501544, | |
| "eval_loss": 0.8533274531364441, | |
| "eval_runtime": 256.963, | |
| "eval_samples_per_second": 158.007, | |
| "eval_steps_per_second": 4.938, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 9.459829997258021, | |
| "grad_norm": 2.864605665206909, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.7574, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 9.459829997258021, | |
| "eval_accuracy": 0.8444638910956694, | |
| "eval_loss": 0.8697899580001831, | |
| "eval_runtime": 255.8551, | |
| "eval_samples_per_second": 158.691, | |
| "eval_steps_per_second": 4.96, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 9.528379489991774, | |
| "grad_norm": 2.5367231369018555, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.7529, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 9.528379489991774, | |
| "eval_accuracy": 0.8451616733099838, | |
| "eval_loss": 0.8582028746604919, | |
| "eval_runtime": 256.6803, | |
| "eval_samples_per_second": 158.181, | |
| "eval_steps_per_second": 4.944, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 9.596928982725528, | |
| "grad_norm": 2.1710877418518066, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7577, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 9.596928982725528, | |
| "eval_accuracy": 0.845897620114414, | |
| "eval_loss": 0.8616137504577637, | |
| "eval_runtime": 255.6697, | |
| "eval_samples_per_second": 158.806, | |
| "eval_steps_per_second": 4.963, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 9.665478475459281, | |
| "grad_norm": 2.400867462158203, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.7554, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 9.665478475459281, | |
| "eval_accuracy": 0.8460372178025299, | |
| "eval_loss": 0.8466119766235352, | |
| "eval_runtime": 256.0692, | |
| "eval_samples_per_second": 158.559, | |
| "eval_steps_per_second": 4.956, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 9.734027968193036, | |
| "grad_norm": 2.6465237140655518, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.7406, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 9.734027968193036, | |
| "eval_accuracy": 0.846062622809469, | |
| "eval_loss": 0.8593913316726685, | |
| "eval_runtime": 257.0976, | |
| "eval_samples_per_second": 157.924, | |
| "eval_steps_per_second": 4.936, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 9.80257746092679, | |
| "grad_norm": 2.72021222114563, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.7543, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 9.80257746092679, | |
| "eval_accuracy": 0.8463149479101548, | |
| "eval_loss": 0.8515172600746155, | |
| "eval_runtime": 256.7199, | |
| "eval_samples_per_second": 158.157, | |
| "eval_steps_per_second": 4.943, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 9.871126953660543, | |
| "grad_norm": 2.639139175415039, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.7506, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 9.871126953660543, | |
| "eval_accuracy": 0.8465278129829884, | |
| "eval_loss": 0.8525589108467102, | |
| "eval_runtime": 258.3778, | |
| "eval_samples_per_second": 157.142, | |
| "eval_steps_per_second": 4.911, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 9.939676446394296, | |
| "grad_norm": 2.2242422103881836, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.7517, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 9.939676446394296, | |
| "eval_accuracy": 0.8470482563336492, | |
| "eval_loss": 0.8529332876205444, | |
| "eval_runtime": 257.3598, | |
| "eval_samples_per_second": 157.764, | |
| "eval_steps_per_second": 4.931, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 10.008225939128051, | |
| "grad_norm": 2.333671808242798, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.7425, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 10.008225939128051, | |
| "eval_accuracy": 0.8467576970668704, | |
| "eval_loss": 0.8543536067008972, | |
| "eval_runtime": 257.3323, | |
| "eval_samples_per_second": 157.78, | |
| "eval_steps_per_second": 4.931, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 10.076775431861805, | |
| "grad_norm": 2.4138877391815186, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.7464, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 10.076775431861805, | |
| "eval_accuracy": 0.8473971965940057, | |
| "eval_loss": 0.8455188870429993, | |
| "eval_runtime": 257.2349, | |
| "eval_samples_per_second": 157.84, | |
| "eval_steps_per_second": 4.933, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 10.145324924595558, | |
| "grad_norm": 2.812563180923462, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.7412, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 10.145324924595558, | |
| "eval_accuracy": 0.8475976099587939, | |
| "eval_loss": 0.8453831076622009, | |
| "eval_runtime": 256.9915, | |
| "eval_samples_per_second": 157.99, | |
| "eval_steps_per_second": 4.938, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 10.213874417329311, | |
| "grad_norm": 2.369260549545288, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.7346, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 10.213874417329311, | |
| "eval_accuracy": 0.8474954397549382, | |
| "eval_loss": 0.8512648344039917, | |
| "eval_runtime": 258.0335, | |
| "eval_samples_per_second": 157.352, | |
| "eval_steps_per_second": 4.918, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 10.282423910063066, | |
| "grad_norm": 2.7622134685516357, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7424, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 10.282423910063066, | |
| "eval_accuracy": 0.8481833714959756, | |
| "eval_loss": 0.8343672156333923, | |
| "eval_runtime": 258.2629, | |
| "eval_samples_per_second": 157.212, | |
| "eval_steps_per_second": 4.914, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 10.35097340279682, | |
| "grad_norm": 2.2065768241882324, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.7364, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 10.35097340279682, | |
| "eval_accuracy": 0.8482113108532771, | |
| "eval_loss": 0.8340145945549011, | |
| "eval_runtime": 258.3206, | |
| "eval_samples_per_second": 157.177, | |
| "eval_steps_per_second": 4.913, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 10.419522895530573, | |
| "grad_norm": 2.186100721359253, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.7409, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 10.419522895530573, | |
| "eval_accuracy": 0.8489686223957396, | |
| "eval_loss": 0.8362465500831604, | |
| "eval_runtime": 256.9187, | |
| "eval_samples_per_second": 158.034, | |
| "eval_steps_per_second": 4.939, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 10.488072388264326, | |
| "grad_norm": 2.706817626953125, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.7353, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 10.488072388264326, | |
| "eval_accuracy": 0.8487501900476949, | |
| "eval_loss": 0.8368015289306641, | |
| "eval_runtime": 257.1636, | |
| "eval_samples_per_second": 157.884, | |
| "eval_steps_per_second": 4.935, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 10.556621880998081, | |
| "grad_norm": 2.523261308670044, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.731, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 10.556621880998081, | |
| "eval_accuracy": 0.8489030226241834, | |
| "eval_loss": 0.8337299823760986, | |
| "eval_runtime": 256.9915, | |
| "eval_samples_per_second": 157.99, | |
| "eval_steps_per_second": 4.938, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 10.625171373731835, | |
| "grad_norm": 2.606250286102295, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.7292, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 10.625171373731835, | |
| "eval_accuracy": 0.8478736538649552, | |
| "eval_loss": 0.8499141335487366, | |
| "eval_runtime": 258.2456, | |
| "eval_samples_per_second": 157.222, | |
| "eval_steps_per_second": 4.914, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 10.693720866465588, | |
| "grad_norm": 2.5361220836639404, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.7359, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 10.693720866465588, | |
| "eval_accuracy": 0.8490765860082904, | |
| "eval_loss": 0.8316646218299866, | |
| "eval_runtime": 258.3157, | |
| "eval_samples_per_second": 157.18, | |
| "eval_steps_per_second": 4.913, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 10.762270359199341, | |
| "grad_norm": 2.3277316093444824, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.7284, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 10.762270359199341, | |
| "eval_accuracy": 0.8495720225312002, | |
| "eval_loss": 0.8365707397460938, | |
| "eval_runtime": 258.2814, | |
| "eval_samples_per_second": 157.201, | |
| "eval_steps_per_second": 4.913, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 10.830819851933096, | |
| "grad_norm": 2.746189832687378, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.7316, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 10.830819851933096, | |
| "eval_accuracy": 0.8500147906280335, | |
| "eval_loss": 0.8251886963844299, | |
| "eval_runtime": 257.4381, | |
| "eval_samples_per_second": 157.716, | |
| "eval_steps_per_second": 4.929, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 10.89936934466685, | |
| "grad_norm": 2.9917354583740234, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.7304, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 10.89936934466685, | |
| "eval_accuracy": 0.8502983624757147, | |
| "eval_loss": 0.8259178996086121, | |
| "eval_runtime": 257.2855, | |
| "eval_samples_per_second": 157.809, | |
| "eval_steps_per_second": 4.932, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 10.967918837400603, | |
| "grad_norm": 2.275324583053589, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7255, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 10.967918837400603, | |
| "eval_accuracy": 0.8505023638996737, | |
| "eval_loss": 0.8250493407249451, | |
| "eval_runtime": 257.2933, | |
| "eval_samples_per_second": 157.804, | |
| "eval_steps_per_second": 4.932, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 11.036468330134356, | |
| "grad_norm": 2.6113440990448, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.7224, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 11.036468330134356, | |
| "eval_accuracy": 0.8507015443771133, | |
| "eval_loss": 0.8299734592437744, | |
| "eval_runtime": 257.4676, | |
| "eval_samples_per_second": 157.698, | |
| "eval_steps_per_second": 4.929, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 11.105017822868112, | |
| "grad_norm": 2.6536693572998047, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.7208, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 11.105017822868112, | |
| "eval_accuracy": 0.8506463117116102, | |
| "eval_loss": 0.8155694007873535, | |
| "eval_runtime": 256.8592, | |
| "eval_samples_per_second": 158.071, | |
| "eval_steps_per_second": 4.94, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 11.173567315601865, | |
| "grad_norm": 2.290782928466797, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.7148, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 11.173567315601865, | |
| "eval_accuracy": 0.8507982915676063, | |
| "eval_loss": 0.8275089859962463, | |
| "eval_runtime": 257.1589, | |
| "eval_samples_per_second": 157.887, | |
| "eval_steps_per_second": 4.935, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 11.242116808335618, | |
| "grad_norm": 2.6533780097961426, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.7193, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 11.242116808335618, | |
| "eval_accuracy": 0.8510864940426184, | |
| "eval_loss": 0.8217721581459045, | |
| "eval_runtime": 258.1645, | |
| "eval_samples_per_second": 157.272, | |
| "eval_steps_per_second": 4.915, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 11.310666301069372, | |
| "grad_norm": 2.6084372997283936, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.7177, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 11.310666301069372, | |
| "eval_accuracy": 0.8508369557490207, | |
| "eval_loss": 0.8289022445678711, | |
| "eval_runtime": 258.043, | |
| "eval_samples_per_second": 157.346, | |
| "eval_steps_per_second": 4.918, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 11.379215793803127, | |
| "grad_norm": 2.2717843055725098, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.7211, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 11.379215793803127, | |
| "eval_accuracy": 0.8514459749572929, | |
| "eval_loss": 0.8198857307434082, | |
| "eval_runtime": 258.3084, | |
| "eval_samples_per_second": 157.184, | |
| "eval_steps_per_second": 4.913, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 11.44776528653688, | |
| "grad_norm": 2.34387469291687, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.7093, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 11.44776528653688, | |
| "eval_accuracy": 0.8511664827348943, | |
| "eval_loss": 0.8272643089294434, | |
| "eval_runtime": 258.1866, | |
| "eval_samples_per_second": 157.258, | |
| "eval_steps_per_second": 4.915, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 11.516314779270633, | |
| "grad_norm": 2.3854498863220215, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.7154, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 11.516314779270633, | |
| "eval_accuracy": 0.8518011662252298, | |
| "eval_loss": 0.8211445212364197, | |
| "eval_runtime": 257.2815, | |
| "eval_samples_per_second": 157.812, | |
| "eval_steps_per_second": 4.932, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 11.584864272004387, | |
| "grad_norm": 2.4457602500915527, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.7178, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 11.584864272004387, | |
| "eval_accuracy": 0.8521036808235916, | |
| "eval_loss": 0.8183203339576721, | |
| "eval_runtime": 258.3778, | |
| "eval_samples_per_second": 157.142, | |
| "eval_steps_per_second": 4.911, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 11.653413764738142, | |
| "grad_norm": 2.5457184314727783, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.716, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 11.653413764738142, | |
| "eval_accuracy": 0.8522289156626506, | |
| "eval_loss": 0.8176619410514832, | |
| "eval_runtime": 256.9844, | |
| "eval_samples_per_second": 157.994, | |
| "eval_steps_per_second": 4.938, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 11.721963257471895, | |
| "grad_norm": 2.855541467666626, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.7081, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 11.721963257471895, | |
| "eval_accuracy": 0.8522235192962014, | |
| "eval_loss": 0.8102879524230957, | |
| "eval_runtime": 257.0844, | |
| "eval_samples_per_second": 157.933, | |
| "eval_steps_per_second": 4.936, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 11.790512750205648, | |
| "grad_norm": 2.44242787361145, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.7112, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 11.790512750205648, | |
| "eval_accuracy": 0.8531087746062566, | |
| "eval_loss": 0.8167855739593506, | |
| "eval_runtime": 258.2921, | |
| "eval_samples_per_second": 157.194, | |
| "eval_steps_per_second": 4.913, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 11.859062242939402, | |
| "grad_norm": 2.559410333633423, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.7089, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 11.859062242939402, | |
| "eval_accuracy": 0.8524401167338927, | |
| "eval_loss": 0.8145312666893005, | |
| "eval_runtime": 258.2263, | |
| "eval_samples_per_second": 157.234, | |
| "eval_steps_per_second": 4.914, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 11.927611735673157, | |
| "grad_norm": 2.3273496627807617, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.7033, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 11.927611735673157, | |
| "eval_accuracy": 0.8526681918388697, | |
| "eval_loss": 0.8153809309005737, | |
| "eval_runtime": 257.196, | |
| "eval_samples_per_second": 157.864, | |
| "eval_steps_per_second": 4.934, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 11.99616122840691, | |
| "grad_norm": 2.354360580444336, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.7029, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 11.99616122840691, | |
| "eval_accuracy": 0.8531567384453895, | |
| "eval_loss": 0.8229334354400635, | |
| "eval_runtime": 257.3002, | |
| "eval_samples_per_second": 157.8, | |
| "eval_steps_per_second": 4.932, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 12.064710721140663, | |
| "grad_norm": 2.4728591442108154, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.7058, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 12.064710721140663, | |
| "eval_accuracy": 0.8537915504584094, | |
| "eval_loss": 0.8140564560890198, | |
| "eval_runtime": 258.2835, | |
| "eval_samples_per_second": 157.199, | |
| "eval_steps_per_second": 4.913, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 12.133260213874417, | |
| "grad_norm": 2.485384702682495, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.7005, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 12.133260213874417, | |
| "eval_accuracy": 0.8534658761375271, | |
| "eval_loss": 0.8151687383651733, | |
| "eval_runtime": 258.2434, | |
| "eval_samples_per_second": 157.224, | |
| "eval_steps_per_second": 4.914, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 12.201809706608172, | |
| "grad_norm": 2.530062198638916, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.6992, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 12.201809706608172, | |
| "eval_accuracy": 0.853826029943314, | |
| "eval_loss": 0.8016021847724915, | |
| "eval_runtime": 258.2461, | |
| "eval_samples_per_second": 157.222, | |
| "eval_steps_per_second": 4.914, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 12.270359199341925, | |
| "grad_norm": 2.5869436264038086, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.7008, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 12.270359199341925, | |
| "eval_accuracy": 0.8535235114498525, | |
| "eval_loss": 0.8112274408340454, | |
| "eval_runtime": 257.5365, | |
| "eval_samples_per_second": 157.655, | |
| "eval_steps_per_second": 4.927, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 12.338908692075679, | |
| "grad_norm": 2.8641934394836426, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6979, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 12.338908692075679, | |
| "eval_accuracy": 0.8538098856943305, | |
| "eval_loss": 0.8109295964241028, | |
| "eval_runtime": 258.3955, | |
| "eval_samples_per_second": 157.131, | |
| "eval_steps_per_second": 4.911, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 12.407458184809432, | |
| "grad_norm": 2.686566114425659, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.6949, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 12.407458184809432, | |
| "eval_accuracy": 0.8543918997918827, | |
| "eval_loss": 0.8125308156013489, | |
| "eval_runtime": 258.1617, | |
| "eval_samples_per_second": 157.274, | |
| "eval_steps_per_second": 4.916, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 12.476007677543187, | |
| "grad_norm": 2.452526569366455, | |
| "learning_rate": 9e-06, | |
| "loss": 0.6946, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 12.476007677543187, | |
| "eval_accuracy": 0.8538440239723134, | |
| "eval_loss": 0.8097832798957825, | |
| "eval_runtime": 258.2364, | |
| "eval_samples_per_second": 157.228, | |
| "eval_steps_per_second": 4.914, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 12.54455717027694, | |
| "grad_norm": 2.463740825653076, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.6939, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 12.54455717027694, | |
| "eval_accuracy": 0.854438946024891, | |
| "eval_loss": 0.7999902963638306, | |
| "eval_runtime": 257.2634, | |
| "eval_samples_per_second": 157.823, | |
| "eval_steps_per_second": 4.933, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 12.613106663010694, | |
| "grad_norm": 2.547820568084717, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.6969, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 12.613106663010694, | |
| "eval_accuracy": 0.8542858590534823, | |
| "eval_loss": 0.8070544004440308, | |
| "eval_runtime": 257.3491, | |
| "eval_samples_per_second": 157.77, | |
| "eval_steps_per_second": 4.931, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 12.681656155744447, | |
| "grad_norm": 2.2731072902679443, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.6967, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 12.681656155744447, | |
| "eval_accuracy": 0.8545510749739543, | |
| "eval_loss": 0.7984638810157776, | |
| "eval_runtime": 258.6739, | |
| "eval_samples_per_second": 156.962, | |
| "eval_steps_per_second": 4.906, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 12.750205648478202, | |
| "grad_norm": 2.242337226867676, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.6944, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 12.750205648478202, | |
| "eval_accuracy": 0.85512864037364, | |
| "eval_loss": 0.7989787459373474, | |
| "eval_runtime": 258.5499, | |
| "eval_samples_per_second": 157.037, | |
| "eval_steps_per_second": 4.908, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 12.818755141211955, | |
| "grad_norm": 2.4914486408233643, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.6885, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 12.818755141211955, | |
| "eval_accuracy": 0.8550956587798725, | |
| "eval_loss": 0.8170965313911438, | |
| "eval_runtime": 258.3692, | |
| "eval_samples_per_second": 157.147, | |
| "eval_steps_per_second": 4.912, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 12.887304633945709, | |
| "grad_norm": 2.1909425258636475, | |
| "learning_rate": 6e-06, | |
| "loss": 0.6897, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 12.887304633945709, | |
| "eval_accuracy": 0.8550768677242255, | |
| "eval_loss": 0.8014948964118958, | |
| "eval_runtime": 258.4993, | |
| "eval_samples_per_second": 157.068, | |
| "eval_steps_per_second": 4.909, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 12.955854126679462, | |
| "grad_norm": 2.7882330417633057, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.7027, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 12.955854126679462, | |
| "eval_accuracy": 0.8551748096354626, | |
| "eval_loss": 0.8074929118156433, | |
| "eval_runtime": 258.4421, | |
| "eval_samples_per_second": 157.103, | |
| "eval_steps_per_second": 4.91, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 13.024403619413217, | |
| "grad_norm": 2.189662218093872, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6926, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 13.024403619413217, | |
| "eval_accuracy": 0.8554257092266915, | |
| "eval_loss": 0.8118977546691895, | |
| "eval_runtime": 257.8701, | |
| "eval_samples_per_second": 157.451, | |
| "eval_steps_per_second": 4.921, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 13.09295311214697, | |
| "grad_norm": 2.4796793460845947, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.697, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 13.09295311214697, | |
| "eval_accuracy": 0.8558198770391428, | |
| "eval_loss": 0.7951220870018005, | |
| "eval_runtime": 258.6851, | |
| "eval_samples_per_second": 156.955, | |
| "eval_steps_per_second": 4.906, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 13.161502604880724, | |
| "grad_norm": 2.475494146347046, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.6814, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 13.161502604880724, | |
| "eval_accuracy": 0.8557974835152675, | |
| "eval_loss": 0.7992942929267883, | |
| "eval_runtime": 258.6884, | |
| "eval_samples_per_second": 156.953, | |
| "eval_steps_per_second": 4.906, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 13.230052097614477, | |
| "grad_norm": 2.662364959716797, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.687, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 13.230052097614477, | |
| "eval_accuracy": 0.8556396210250248, | |
| "eval_loss": 0.7970269322395325, | |
| "eval_runtime": 257.7318, | |
| "eval_samples_per_second": 157.536, | |
| "eval_steps_per_second": 4.924, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 13.298601590348232, | |
| "grad_norm": 2.5556256771087646, | |
| "learning_rate": 3e-06, | |
| "loss": 0.6956, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 13.298601590348232, | |
| "eval_accuracy": 0.8560198209787908, | |
| "eval_loss": 0.7952587008476257, | |
| "eval_runtime": 258.556, | |
| "eval_samples_per_second": 157.034, | |
| "eval_steps_per_second": 4.908, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 13.367151083081986, | |
| "grad_norm": 2.327164888381958, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.6821, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 13.367151083081986, | |
| "eval_accuracy": 0.855291832818916, | |
| "eval_loss": 0.8007811307907104, | |
| "eval_runtime": 257.5846, | |
| "eval_samples_per_second": 157.626, | |
| "eval_steps_per_second": 4.927, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 13.435700575815739, | |
| "grad_norm": 2.408548593521118, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.6846, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 13.435700575815739, | |
| "eval_accuracy": 0.8561709926629, | |
| "eval_loss": 0.7897204756736755, | |
| "eval_runtime": 258.6709, | |
| "eval_samples_per_second": 156.964, | |
| "eval_steps_per_second": 4.906, | |
| "step": 98000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.262679236758733e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |