| { | |
| "best_metric": 0.8528650403022766, | |
| "best_model_checkpoint": "./model_fine-tune/glot/mbert/dan-Latn/checkpoint-98500", | |
| "epoch": 10.286131996658312, | |
| "eval_steps": 500, | |
| "global_step": 98500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.052213868003341685, | |
| "grad_norm": 4.64046049118042, | |
| "learning_rate": 9.95e-05, | |
| "loss": 1.7666, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.052213868003341685, | |
| "eval_accuracy": 0.6961985245587479, | |
| "eval_loss": 1.741060733795166, | |
| "eval_runtime": 361.3155, | |
| "eval_samples_per_second": 175.589, | |
| "eval_steps_per_second": 5.488, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.10442773600668337, | |
| "grad_norm": 3.5290920734405518, | |
| "learning_rate": 9.900000000000001e-05, | |
| "loss": 1.6122, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.10442773600668337, | |
| "eval_accuracy": 0.7084748820521898, | |
| "eval_loss": 1.6247525215148926, | |
| "eval_runtime": 360.2132, | |
| "eval_samples_per_second": 176.126, | |
| "eval_steps_per_second": 5.505, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "grad_norm": 3.951317071914673, | |
| "learning_rate": 9.850000000000001e-05, | |
| "loss": 1.5487, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "eval_accuracy": 0.718550082660156, | |
| "eval_loss": 1.553884506225586, | |
| "eval_runtime": 361.2942, | |
| "eval_samples_per_second": 175.599, | |
| "eval_steps_per_second": 5.489, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.20885547201336674, | |
| "grad_norm": 3.799274444580078, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.4946, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.20885547201336674, | |
| "eval_accuracy": 0.723134633812131, | |
| "eval_loss": 1.5116215944290161, | |
| "eval_runtime": 360.3813, | |
| "eval_samples_per_second": 176.044, | |
| "eval_steps_per_second": 5.503, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.26106934001670845, | |
| "grad_norm": 3.499929428100586, | |
| "learning_rate": 9.75e-05, | |
| "loss": 1.4573, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.26106934001670845, | |
| "eval_accuracy": 0.7287729394802086, | |
| "eval_loss": 1.4748759269714355, | |
| "eval_runtime": 362.14, | |
| "eval_samples_per_second": 175.189, | |
| "eval_steps_per_second": 5.476, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "grad_norm": 3.6163783073425293, | |
| "learning_rate": 9.7e-05, | |
| "loss": 1.4332, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "eval_accuracy": 0.7334675822291515, | |
| "eval_loss": 1.4382221698760986, | |
| "eval_runtime": 362.0967, | |
| "eval_samples_per_second": 175.21, | |
| "eval_steps_per_second": 5.476, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3654970760233918, | |
| "grad_norm": 3.382498264312744, | |
| "learning_rate": 9.65e-05, | |
| "loss": 1.3914, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3654970760233918, | |
| "eval_accuracy": 0.7367988898749408, | |
| "eval_loss": 1.4078316688537598, | |
| "eval_runtime": 362.0182, | |
| "eval_samples_per_second": 175.248, | |
| "eval_steps_per_second": 5.478, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.4177109440267335, | |
| "grad_norm": 3.2594001293182373, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.3843, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4177109440267335, | |
| "eval_accuracy": 0.7404781561454725, | |
| "eval_loss": 1.37874174118042, | |
| "eval_runtime": 362.2257, | |
| "eval_samples_per_second": 175.148, | |
| "eval_steps_per_second": 5.474, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "grad_norm": 3.717684268951416, | |
| "learning_rate": 9.55e-05, | |
| "loss": 1.3628, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "eval_accuracy": 0.7430362394212798, | |
| "eval_loss": 1.373829960823059, | |
| "eval_runtime": 362.0885, | |
| "eval_samples_per_second": 175.214, | |
| "eval_steps_per_second": 5.477, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.5221386800334169, | |
| "grad_norm": 3.67509126663208, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.356, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5221386800334169, | |
| "eval_accuracy": 0.7465132147816155, | |
| "eval_loss": 1.343856930732727, | |
| "eval_runtime": 362.013, | |
| "eval_samples_per_second": 175.251, | |
| "eval_steps_per_second": 5.478, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5743525480367586, | |
| "grad_norm": 2.995349407196045, | |
| "learning_rate": 9.449999999999999e-05, | |
| "loss": 1.3275, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5743525480367586, | |
| "eval_accuracy": 0.747906382026035, | |
| "eval_loss": 1.3185055255889893, | |
| "eval_runtime": 363.6082, | |
| "eval_samples_per_second": 174.482, | |
| "eval_steps_per_second": 5.454, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.6265664160401002, | |
| "grad_norm": 3.0733938217163086, | |
| "learning_rate": 9.4e-05, | |
| "loss": 1.3192, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6265664160401002, | |
| "eval_accuracy": 0.7498967127579244, | |
| "eval_loss": 1.3267806768417358, | |
| "eval_runtime": 362.4596, | |
| "eval_samples_per_second": 175.035, | |
| "eval_steps_per_second": 5.471, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.6787802840434419, | |
| "grad_norm": 3.0954840183258057, | |
| "learning_rate": 9.350000000000001e-05, | |
| "loss": 1.3123, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6787802840434419, | |
| "eval_accuracy": 0.7517548935891929, | |
| "eval_loss": 1.3006523847579956, | |
| "eval_runtime": 363.2093, | |
| "eval_samples_per_second": 174.673, | |
| "eval_steps_per_second": 5.46, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.7309941520467836, | |
| "grad_norm": 3.1289381980895996, | |
| "learning_rate": 9.300000000000001e-05, | |
| "loss": 1.2942, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7309941520467836, | |
| "eval_accuracy": 0.7542296938604579, | |
| "eval_loss": 1.2929292917251587, | |
| "eval_runtime": 362.9614, | |
| "eval_samples_per_second": 174.793, | |
| "eval_steps_per_second": 5.463, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.7832080200501254, | |
| "grad_norm": 3.440656900405884, | |
| "learning_rate": 9.250000000000001e-05, | |
| "loss": 1.2812, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7832080200501254, | |
| "eval_accuracy": 0.756235706841684, | |
| "eval_loss": 1.2668538093566895, | |
| "eval_runtime": 362.1254, | |
| "eval_samples_per_second": 175.196, | |
| "eval_steps_per_second": 5.476, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.835421888053467, | |
| "grad_norm": 4.0889458656311035, | |
| "learning_rate": 9.200000000000001e-05, | |
| "loss": 1.2708, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.835421888053467, | |
| "eval_accuracy": 0.7571290622439745, | |
| "eval_loss": 1.283825159072876, | |
| "eval_runtime": 362.2084, | |
| "eval_samples_per_second": 175.156, | |
| "eval_steps_per_second": 5.475, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8876357560568087, | |
| "grad_norm": 2.77811598777771, | |
| "learning_rate": 9.15e-05, | |
| "loss": 1.2759, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8876357560568087, | |
| "eval_accuracy": 0.7586107082301347, | |
| "eval_loss": 1.2615000009536743, | |
| "eval_runtime": 362.3765, | |
| "eval_samples_per_second": 175.075, | |
| "eval_steps_per_second": 5.472, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.9398496240601504, | |
| "grad_norm": 3.350158929824829, | |
| "learning_rate": 9.1e-05, | |
| "loss": 1.257, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9398496240601504, | |
| "eval_accuracy": 0.7595859575845869, | |
| "eval_loss": 1.250738501548767, | |
| "eval_runtime": 362.5674, | |
| "eval_samples_per_second": 174.983, | |
| "eval_steps_per_second": 5.469, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.9920634920634921, | |
| "grad_norm": 3.2566370964050293, | |
| "learning_rate": 9.05e-05, | |
| "loss": 1.2437, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9920634920634921, | |
| "eval_accuracy": 0.7619539601093824, | |
| "eval_loss": 1.248534083366394, | |
| "eval_runtime": 362.5746, | |
| "eval_samples_per_second": 174.979, | |
| "eval_steps_per_second": 5.469, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.0442773600668338, | |
| "grad_norm": 3.279320001602173, | |
| "learning_rate": 9e-05, | |
| "loss": 1.2333, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0442773600668338, | |
| "eval_accuracy": 0.762933591347759, | |
| "eval_loss": 1.2348397970199585, | |
| "eval_runtime": 362.8313, | |
| "eval_samples_per_second": 174.855, | |
| "eval_steps_per_second": 5.465, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.0964912280701755, | |
| "grad_norm": 3.002204418182373, | |
| "learning_rate": 8.950000000000001e-05, | |
| "loss": 1.2204, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0964912280701755, | |
| "eval_accuracy": 0.7645752968362766, | |
| "eval_loss": 1.2361088991165161, | |
| "eval_runtime": 362.7427, | |
| "eval_samples_per_second": 174.898, | |
| "eval_steps_per_second": 5.467, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.1487050960735172, | |
| "grad_norm": 4.058037281036377, | |
| "learning_rate": 8.900000000000001e-05, | |
| "loss": 1.2116, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.1487050960735172, | |
| "eval_accuracy": 0.7654839256801539, | |
| "eval_loss": 1.2149455547332764, | |
| "eval_runtime": 362.0743, | |
| "eval_samples_per_second": 175.221, | |
| "eval_steps_per_second": 5.477, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.2009189640768587, | |
| "grad_norm": 3.364161968231201, | |
| "learning_rate": 8.850000000000001e-05, | |
| "loss": 1.2083, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2009189640768587, | |
| "eval_accuracy": 0.7664035053683952, | |
| "eval_loss": 1.2015998363494873, | |
| "eval_runtime": 362.482, | |
| "eval_samples_per_second": 175.024, | |
| "eval_steps_per_second": 5.471, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.2531328320802004, | |
| "grad_norm": 2.995591402053833, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.2056, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.2531328320802004, | |
| "eval_accuracy": 0.76765140727518, | |
| "eval_loss": 1.1961411237716675, | |
| "eval_runtime": 363.557, | |
| "eval_samples_per_second": 174.506, | |
| "eval_steps_per_second": 5.454, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.3053467000835421, | |
| "grad_norm": 3.010730028152466, | |
| "learning_rate": 8.75e-05, | |
| "loss": 1.1917, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.3053467000835421, | |
| "eval_accuracy": 0.7685342108740758, | |
| "eval_loss": 1.1992570161819458, | |
| "eval_runtime": 362.5156, | |
| "eval_samples_per_second": 175.008, | |
| "eval_steps_per_second": 5.47, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.3575605680868839, | |
| "grad_norm": 3.052395820617676, | |
| "learning_rate": 8.7e-05, | |
| "loss": 1.1905, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.3575605680868839, | |
| "eval_accuracy": 0.7699597470201401, | |
| "eval_loss": 1.2011651992797852, | |
| "eval_runtime": 361.326, | |
| "eval_samples_per_second": 175.584, | |
| "eval_steps_per_second": 5.488, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.4097744360902256, | |
| "grad_norm": 2.883955717086792, | |
| "learning_rate": 8.65e-05, | |
| "loss": 1.1839, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.4097744360902256, | |
| "eval_accuracy": 0.7708384570352297, | |
| "eval_loss": 1.1753507852554321, | |
| "eval_runtime": 362.9321, | |
| "eval_samples_per_second": 174.807, | |
| "eval_steps_per_second": 5.464, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.4619883040935673, | |
| "grad_norm": 2.7026546001434326, | |
| "learning_rate": 8.6e-05, | |
| "loss": 1.1724, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.4619883040935673, | |
| "eval_accuracy": 0.7709277011489558, | |
| "eval_loss": 1.168317198753357, | |
| "eval_runtime": 362.8518, | |
| "eval_samples_per_second": 174.845, | |
| "eval_steps_per_second": 5.465, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.514202172096909, | |
| "grad_norm": 2.8576297760009766, | |
| "learning_rate": 8.55e-05, | |
| "loss": 1.1758, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.514202172096909, | |
| "eval_accuracy": 0.7720082438542706, | |
| "eval_loss": 1.1711252927780151, | |
| "eval_runtime": 362.6486, | |
| "eval_samples_per_second": 174.943, | |
| "eval_steps_per_second": 5.468, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.5664160401002505, | |
| "grad_norm": 3.314800500869751, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.1656, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.5664160401002505, | |
| "eval_accuracy": 0.7731014057112545, | |
| "eval_loss": NaN, | |
| "eval_runtime": 362.4489, | |
| "eval_samples_per_second": 175.04, | |
| "eval_steps_per_second": 5.471, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.6186299081035922, | |
| "grad_norm": 2.740029811859131, | |
| "learning_rate": 8.450000000000001e-05, | |
| "loss": 1.1659, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.6186299081035922, | |
| "eval_accuracy": 0.7745773535784676, | |
| "eval_loss": 1.1655621528625488, | |
| "eval_runtime": 362.0759, | |
| "eval_samples_per_second": 175.22, | |
| "eval_steps_per_second": 5.477, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.670843776106934, | |
| "grad_norm": 2.9377198219299316, | |
| "learning_rate": 8.4e-05, | |
| "loss": 1.1512, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.670843776106934, | |
| "eval_accuracy": 0.7745256599790964, | |
| "eval_loss": 1.1678109169006348, | |
| "eval_runtime": 362.3938, | |
| "eval_samples_per_second": 175.066, | |
| "eval_steps_per_second": 5.472, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.7230576441102756, | |
| "grad_norm": 3.2804691791534424, | |
| "learning_rate": 8.35e-05, | |
| "loss": 1.1394, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.7230576441102756, | |
| "eval_accuracy": 0.7751363378115868, | |
| "eval_loss": 1.1540894508361816, | |
| "eval_runtime": 362.0672, | |
| "eval_samples_per_second": 175.224, | |
| "eval_steps_per_second": 5.477, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.7752715121136173, | |
| "grad_norm": 3.486011028289795, | |
| "learning_rate": 8.3e-05, | |
| "loss": 1.1382, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.7752715121136173, | |
| "eval_accuracy": 0.7765838022142181, | |
| "eval_loss": 1.1509926319122314, | |
| "eval_runtime": 362.2521, | |
| "eval_samples_per_second": 175.135, | |
| "eval_steps_per_second": 5.474, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.827485380116959, | |
| "grad_norm": 2.56192684173584, | |
| "learning_rate": 8.25e-05, | |
| "loss": 1.1448, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.827485380116959, | |
| "eval_accuracy": 0.7774433499499322, | |
| "eval_loss": 1.150856375694275, | |
| "eval_runtime": 361.437, | |
| "eval_samples_per_second": 175.53, | |
| "eval_steps_per_second": 5.486, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "grad_norm": 2.9113447666168213, | |
| "learning_rate": 8.2e-05, | |
| "loss": 1.1376, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.8796992481203008, | |
| "eval_accuracy": 0.7779668095978695, | |
| "eval_loss": 1.1359950304031372, | |
| "eval_runtime": 362.1516, | |
| "eval_samples_per_second": 175.184, | |
| "eval_steps_per_second": 5.476, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.9319131161236425, | |
| "grad_norm": 2.8685195446014404, | |
| "learning_rate": 8.15e-05, | |
| "loss": 1.1306, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.9319131161236425, | |
| "eval_accuracy": 0.7786356827454954, | |
| "eval_loss": 1.1356443166732788, | |
| "eval_runtime": 367.0737, | |
| "eval_samples_per_second": 172.835, | |
| "eval_steps_per_second": 5.402, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "grad_norm": 3.4618313312530518, | |
| "learning_rate": 8.1e-05, | |
| "loss": 1.1379, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.9841269841269842, | |
| "eval_accuracy": 0.7796485696787286, | |
| "eval_loss": 1.1357932090759277, | |
| "eval_runtime": 367.2471, | |
| "eval_samples_per_second": 172.753, | |
| "eval_steps_per_second": 5.4, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.036340852130326, | |
| "grad_norm": 3.1010448932647705, | |
| "learning_rate": 8.05e-05, | |
| "loss": 1.109, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.036340852130326, | |
| "eval_accuracy": 0.7807163576099158, | |
| "eval_loss": 1.1260879039764404, | |
| "eval_runtime": 367.0489, | |
| "eval_samples_per_second": 172.846, | |
| "eval_steps_per_second": 5.403, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.0885547201336676, | |
| "grad_norm": 3.1995084285736084, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1047, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.0885547201336676, | |
| "eval_accuracy": 0.7806008208598338, | |
| "eval_loss": 1.1227824687957764, | |
| "eval_runtime": 367.4627, | |
| "eval_samples_per_second": 172.652, | |
| "eval_steps_per_second": 5.396, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.1407685881370093, | |
| "grad_norm": 2.741629123687744, | |
| "learning_rate": 7.950000000000001e-05, | |
| "loss": 1.1042, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.1407685881370093, | |
| "eval_accuracy": 0.7819552122122398, | |
| "eval_loss": 1.1106891632080078, | |
| "eval_runtime": 369.2375, | |
| "eval_samples_per_second": 171.822, | |
| "eval_steps_per_second": 5.371, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.192982456140351, | |
| "grad_norm": 3.0432229042053223, | |
| "learning_rate": 7.900000000000001e-05, | |
| "loss": 1.1071, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.192982456140351, | |
| "eval_accuracy": 0.7825761226202902, | |
| "eval_loss": 1.105894923210144, | |
| "eval_runtime": 368.3863, | |
| "eval_samples_per_second": 172.219, | |
| "eval_steps_per_second": 5.383, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.2451963241436927, | |
| "grad_norm": 2.60355806350708, | |
| "learning_rate": 7.850000000000001e-05, | |
| "loss": 1.1, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.2451963241436927, | |
| "eval_accuracy": 0.7824829036453502, | |
| "eval_loss": 1.1067482233047485, | |
| "eval_runtime": 369.3294, | |
| "eval_samples_per_second": 171.779, | |
| "eval_steps_per_second": 5.369, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.2974101921470345, | |
| "grad_norm": 3.290499448776245, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.094, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.2974101921470345, | |
| "eval_accuracy": 0.7835772317290801, | |
| "eval_loss": 1.1064658164978027, | |
| "eval_runtime": 371.2345, | |
| "eval_samples_per_second": 170.897, | |
| "eval_steps_per_second": 5.342, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.3496240601503757, | |
| "grad_norm": 2.749668598175049, | |
| "learning_rate": 7.75e-05, | |
| "loss": 1.0868, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.3496240601503757, | |
| "eval_accuracy": 0.7844280402341692, | |
| "eval_loss": 1.1136114597320557, | |
| "eval_runtime": 370.2159, | |
| "eval_samples_per_second": 171.368, | |
| "eval_steps_per_second": 5.356, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.4018379281537174, | |
| "grad_norm": 2.553410768508911, | |
| "learning_rate": 7.7e-05, | |
| "loss": 1.0938, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.4018379281537174, | |
| "eval_accuracy": 0.7854044343476029, | |
| "eval_loss": 1.0965579748153687, | |
| "eval_runtime": 370.6834, | |
| "eval_samples_per_second": 171.151, | |
| "eval_steps_per_second": 5.35, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.454051796157059, | |
| "grad_norm": 2.7348687648773193, | |
| "learning_rate": 7.65e-05, | |
| "loss": 1.0881, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.454051796157059, | |
| "eval_accuracy": 0.7850364727467058, | |
| "eval_loss": 1.1107020378112793, | |
| "eval_runtime": 369.6648, | |
| "eval_samples_per_second": 171.623, | |
| "eval_steps_per_second": 5.364, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.506265664160401, | |
| "grad_norm": 2.639407157897949, | |
| "learning_rate": 7.6e-05, | |
| "loss": 1.0885, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.506265664160401, | |
| "eval_accuracy": 0.7867807742632953, | |
| "eval_loss": 1.0890835523605347, | |
| "eval_runtime": 369.0693, | |
| "eval_samples_per_second": 171.9, | |
| "eval_steps_per_second": 5.373, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.5584795321637426, | |
| "grad_norm": 3.1823441982269287, | |
| "learning_rate": 7.55e-05, | |
| "loss": 1.077, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.5584795321637426, | |
| "eval_accuracy": 0.7863467916817448, | |
| "eval_loss": 1.0876108407974243, | |
| "eval_runtime": 368.1359, | |
| "eval_samples_per_second": 172.336, | |
| "eval_steps_per_second": 5.387, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.6106934001670843, | |
| "grad_norm": 2.6731505393981934, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.078, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.6106934001670843, | |
| "eval_accuracy": 0.7872558672911878, | |
| "eval_loss": 1.0856846570968628, | |
| "eval_runtime": 367.5868, | |
| "eval_samples_per_second": 172.593, | |
| "eval_steps_per_second": 5.395, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.662907268170426, | |
| "grad_norm": 2.8799052238464355, | |
| "learning_rate": 7.450000000000001e-05, | |
| "loss": 1.0759, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.662907268170426, | |
| "eval_accuracy": 0.788363306685935, | |
| "eval_loss": 1.080370545387268, | |
| "eval_runtime": 368.0395, | |
| "eval_samples_per_second": 172.381, | |
| "eval_steps_per_second": 5.388, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.7151211361737677, | |
| "grad_norm": 4.7506208419799805, | |
| "learning_rate": 7.4e-05, | |
| "loss": 1.072, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.7151211361737677, | |
| "eval_accuracy": 0.7885482778687157, | |
| "eval_loss": 1.08516526222229, | |
| "eval_runtime": 366.0953, | |
| "eval_samples_per_second": 173.296, | |
| "eval_steps_per_second": 5.417, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.7673350041771094, | |
| "grad_norm": 2.411212682723999, | |
| "learning_rate": 7.35e-05, | |
| "loss": 1.0655, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.7673350041771094, | |
| "eval_accuracy": 0.7889261779231521, | |
| "eval_loss": 1.0773206949234009, | |
| "eval_runtime": 366.4448, | |
| "eval_samples_per_second": 173.131, | |
| "eval_steps_per_second": 5.411, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.819548872180451, | |
| "grad_norm": 2.6230037212371826, | |
| "learning_rate": 7.3e-05, | |
| "loss": 1.0747, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.819548872180451, | |
| "eval_accuracy": 0.7891164707308205, | |
| "eval_loss": 1.0783028602600098, | |
| "eval_runtime": 367.8248, | |
| "eval_samples_per_second": 172.482, | |
| "eval_steps_per_second": 5.391, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.871762740183793, | |
| "grad_norm": 2.9123287200927734, | |
| "learning_rate": 7.25e-05, | |
| "loss": 1.0691, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.871762740183793, | |
| "eval_accuracy": 0.7904697432780537, | |
| "eval_loss": 1.0683778524398804, | |
| "eval_runtime": 365.814, | |
| "eval_samples_per_second": 173.43, | |
| "eval_steps_per_second": 5.421, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.9239766081871346, | |
| "grad_norm": 2.716794967651367, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.0581, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.9239766081871346, | |
| "eval_accuracy": 0.7907833141699024, | |
| "eval_loss": 1.0698238611221313, | |
| "eval_runtime": 368.4835, | |
| "eval_samples_per_second": 172.173, | |
| "eval_steps_per_second": 5.382, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.9761904761904763, | |
| "grad_norm": 2.900315761566162, | |
| "learning_rate": 7.15e-05, | |
| "loss": 1.0576, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.9761904761904763, | |
| "eval_accuracy": 0.7912189924393805, | |
| "eval_loss": 1.0527312755584717, | |
| "eval_runtime": 367.3342, | |
| "eval_samples_per_second": 172.712, | |
| "eval_steps_per_second": 5.398, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 3.028404344193818, | |
| "grad_norm": 2.733102321624756, | |
| "learning_rate": 7.1e-05, | |
| "loss": 1.0491, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.028404344193818, | |
| "eval_accuracy": 0.7912027755966868, | |
| "eval_loss": 1.0649120807647705, | |
| "eval_runtime": 367.5821, | |
| "eval_samples_per_second": 172.595, | |
| "eval_steps_per_second": 5.395, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 3.0806182121971597, | |
| "grad_norm": 2.990572452545166, | |
| "learning_rate": 7.05e-05, | |
| "loss": 1.0429, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.0806182121971597, | |
| "eval_accuracy": 0.791866931892446, | |
| "eval_loss": 1.065590262413025, | |
| "eval_runtime": 367.6583, | |
| "eval_samples_per_second": 172.56, | |
| "eval_steps_per_second": 5.394, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 3.1328320802005014, | |
| "grad_norm": 2.6242377758026123, | |
| "learning_rate": 7e-05, | |
| "loss": 1.0416, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.1328320802005014, | |
| "eval_accuracy": 0.793420728495659, | |
| "eval_loss": 1.056522250175476, | |
| "eval_runtime": 367.1286, | |
| "eval_samples_per_second": 172.809, | |
| "eval_steps_per_second": 5.401, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 3.185045948203843, | |
| "grad_norm": 2.6089892387390137, | |
| "learning_rate": 6.95e-05, | |
| "loss": 1.0388, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.185045948203843, | |
| "eval_accuracy": 0.792953820707681, | |
| "eval_loss": 1.0602775812149048, | |
| "eval_runtime": 367.5358, | |
| "eval_samples_per_second": 172.617, | |
| "eval_steps_per_second": 5.395, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 3.2372598162071844, | |
| "grad_norm": 2.6940767765045166, | |
| "learning_rate": 6.9e-05, | |
| "loss": 1.0351, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.2372598162071844, | |
| "eval_accuracy": 0.7933434972497504, | |
| "eval_loss": 1.0466923713684082, | |
| "eval_runtime": 367.3002, | |
| "eval_samples_per_second": 172.728, | |
| "eval_steps_per_second": 5.399, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "grad_norm": 2.5024611949920654, | |
| "learning_rate": 6.850000000000001e-05, | |
| "loss": 1.0364, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.2894736842105265, | |
| "eval_accuracy": 0.7938457165216277, | |
| "eval_loss": 1.0515408515930176, | |
| "eval_runtime": 367.3309, | |
| "eval_samples_per_second": 172.713, | |
| "eval_steps_per_second": 5.398, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 3.341687552213868, | |
| "grad_norm": 2.594076156616211, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.0352, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.341687552213868, | |
| "eval_accuracy": 0.7949145678709858, | |
| "eval_loss": 1.0443620681762695, | |
| "eval_runtime": 367.752, | |
| "eval_samples_per_second": 172.516, | |
| "eval_steps_per_second": 5.392, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 3.3939014202172095, | |
| "grad_norm": 4.887377738952637, | |
| "learning_rate": 6.750000000000001e-05, | |
| "loss": 1.0269, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.3939014202172095, | |
| "eval_accuracy": 0.7945826018068661, | |
| "eval_loss": 1.0488784313201904, | |
| "eval_runtime": 368.3885, | |
| "eval_samples_per_second": 172.218, | |
| "eval_steps_per_second": 5.383, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 3.4461152882205512, | |
| "grad_norm": 3.1437246799468994, | |
| "learning_rate": 6.7e-05, | |
| "loss": 1.0293, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.4461152882205512, | |
| "eval_accuracy": 0.795481144785727, | |
| "eval_loss": 1.0364991426467896, | |
| "eval_runtime": 366.6755, | |
| "eval_samples_per_second": 173.022, | |
| "eval_steps_per_second": 5.408, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 3.498329156223893, | |
| "grad_norm": 2.490694761276245, | |
| "learning_rate": 6.65e-05, | |
| "loss": 1.0179, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.498329156223893, | |
| "eval_accuracy": 0.7964686907750312, | |
| "eval_loss": 1.043065071105957, | |
| "eval_runtime": 367.4767, | |
| "eval_samples_per_second": 172.645, | |
| "eval_steps_per_second": 5.396, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 3.5505430242272347, | |
| "grad_norm": 2.9056098461151123, | |
| "learning_rate": 6.6e-05, | |
| "loss": 1.0147, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.5505430242272347, | |
| "eval_accuracy": 0.7962407550483477, | |
| "eval_loss": NaN, | |
| "eval_runtime": 367.3507, | |
| "eval_samples_per_second": 172.704, | |
| "eval_steps_per_second": 5.398, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 3.6027568922305764, | |
| "grad_norm": 2.655351161956787, | |
| "learning_rate": 6.55e-05, | |
| "loss": 1.0112, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.6027568922305764, | |
| "eval_accuracy": 0.7970424400295055, | |
| "eval_loss": 1.033272624015808, | |
| "eval_runtime": 367.4604, | |
| "eval_samples_per_second": 172.653, | |
| "eval_steps_per_second": 5.397, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 3.654970760233918, | |
| "grad_norm": 2.922175168991089, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.0145, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.654970760233918, | |
| "eval_accuracy": 0.7967277004610398, | |
| "eval_loss": 1.0407856702804565, | |
| "eval_runtime": 367.6311, | |
| "eval_samples_per_second": 172.572, | |
| "eval_steps_per_second": 5.394, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 3.70718462823726, | |
| "grad_norm": 2.7691986560821533, | |
| "learning_rate": 6.450000000000001e-05, | |
| "loss": 1.0133, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.70718462823726, | |
| "eval_accuracy": 0.7978057552609327, | |
| "eval_loss": 1.0315524339675903, | |
| "eval_runtime": 368.5127, | |
| "eval_samples_per_second": 172.16, | |
| "eval_steps_per_second": 5.381, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "grad_norm": 2.718721389770508, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.0123, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.7593984962406015, | |
| "eval_accuracy": 0.7984869599976302, | |
| "eval_loss": 1.0284632444381714, | |
| "eval_runtime": 367.5592, | |
| "eval_samples_per_second": 172.606, | |
| "eval_steps_per_second": 5.395, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 3.8116123642439432, | |
| "grad_norm": 2.4457788467407227, | |
| "learning_rate": 6.35e-05, | |
| "loss": 1.008, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.8116123642439432, | |
| "eval_accuracy": 0.7985738639575599, | |
| "eval_loss": 1.0290788412094116, | |
| "eval_runtime": 367.5585, | |
| "eval_samples_per_second": 172.607, | |
| "eval_steps_per_second": 5.395, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 3.863826232247285, | |
| "grad_norm": 3.0719549655914307, | |
| "learning_rate": 6.3e-05, | |
| "loss": 0.9995, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.863826232247285, | |
| "eval_accuracy": 0.7990341348986376, | |
| "eval_loss": 1.0225682258605957, | |
| "eval_runtime": 367.5861, | |
| "eval_samples_per_second": 172.594, | |
| "eval_steps_per_second": 5.395, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.9160401002506267, | |
| "grad_norm": 2.502732038497925, | |
| "learning_rate": 6.25e-05, | |
| "loss": 1.0059, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.9160401002506267, | |
| "eval_accuracy": 0.7990824767003194, | |
| "eval_loss": 1.0225658416748047, | |
| "eval_runtime": 368.4364, | |
| "eval_samples_per_second": 172.195, | |
| "eval_steps_per_second": 5.382, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "grad_norm": 2.615722894668579, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.9908, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.9682539682539684, | |
| "eval_accuracy": 0.8004074602540703, | |
| "eval_loss": NaN, | |
| "eval_runtime": 367.3504, | |
| "eval_samples_per_second": 172.704, | |
| "eval_steps_per_second": 5.398, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 4.02046783625731, | |
| "grad_norm": 2.6223480701446533, | |
| "learning_rate": 6.15e-05, | |
| "loss": 1.0019, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.02046783625731, | |
| "eval_accuracy": 0.8002175287186218, | |
| "eval_loss": 1.0221267938613892, | |
| "eval_runtime": 367.5886, | |
| "eval_samples_per_second": 172.592, | |
| "eval_steps_per_second": 5.395, | |
| "step": 38500 | |
| }, | |
| { | |
| "epoch": 4.072681704260652, | |
| "grad_norm": 7.730719089508057, | |
| "learning_rate": 6.1e-05, | |
| "loss": 0.9859, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.072681704260652, | |
| "eval_accuracy": 0.8004946972174642, | |
| "eval_loss": 1.0225615501403809, | |
| "eval_runtime": 367.5118, | |
| "eval_samples_per_second": 172.628, | |
| "eval_steps_per_second": 5.396, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 4.124895572263993, | |
| "grad_norm": 2.6854565143585205, | |
| "learning_rate": 6.05e-05, | |
| "loss": 0.9899, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.124895572263993, | |
| "eval_accuracy": 0.8006603707418583, | |
| "eval_loss": 1.0182828903198242, | |
| "eval_runtime": 367.2006, | |
| "eval_samples_per_second": 172.775, | |
| "eval_steps_per_second": 5.4, | |
| "step": 39500 | |
| }, | |
| { | |
| "epoch": 4.177109440267335, | |
| "grad_norm": 2.871232509613037, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9769, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.177109440267335, | |
| "eval_accuracy": 0.8012399052534226, | |
| "eval_loss": 1.007629632949829, | |
| "eval_runtime": 367.698, | |
| "eval_samples_per_second": 172.541, | |
| "eval_steps_per_second": 5.393, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 4.2293233082706765, | |
| "grad_norm": 2.7055728435516357, | |
| "learning_rate": 5.95e-05, | |
| "loss": 0.9862, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.2293233082706765, | |
| "eval_accuracy": 0.8023664324313652, | |
| "eval_loss": 1.0073480606079102, | |
| "eval_runtime": 367.5166, | |
| "eval_samples_per_second": 172.626, | |
| "eval_steps_per_second": 5.396, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 4.281537176274019, | |
| "grad_norm": 2.628458261489868, | |
| "learning_rate": 5.9e-05, | |
| "loss": 0.9784, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.281537176274019, | |
| "eval_accuracy": 0.8022199523230344, | |
| "eval_loss": 1.0004937648773193, | |
| "eval_runtime": 368.0641, | |
| "eval_samples_per_second": 172.369, | |
| "eval_steps_per_second": 5.388, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 4.33375104427736, | |
| "grad_norm": 2.5863466262817383, | |
| "learning_rate": 5.85e-05, | |
| "loss": 0.974, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.33375104427736, | |
| "eval_accuracy": 0.8025649357830319, | |
| "eval_loss": 1.0010262727737427, | |
| "eval_runtime": 367.7203, | |
| "eval_samples_per_second": 172.531, | |
| "eval_steps_per_second": 5.393, | |
| "step": 41500 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "grad_norm": 2.748701333999634, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.9828, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "eval_accuracy": 0.8030370401871146, | |
| "eval_loss": 0.998631477355957, | |
| "eval_runtime": 367.8751, | |
| "eval_samples_per_second": 172.458, | |
| "eval_steps_per_second": 5.39, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 4.438178780284043, | |
| "grad_norm": 2.7004311084747314, | |
| "learning_rate": 5.7499999999999995e-05, | |
| "loss": 0.9863, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.438178780284043, | |
| "eval_accuracy": 0.8032171565331311, | |
| "eval_loss": 1.0026530027389526, | |
| "eval_runtime": 367.6912, | |
| "eval_samples_per_second": 172.544, | |
| "eval_steps_per_second": 5.393, | |
| "step": 42500 | |
| }, | |
| { | |
| "epoch": 4.4903926482873855, | |
| "grad_norm": 2.4777750968933105, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.9725, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.4903926482873855, | |
| "eval_accuracy": 0.8036803447902678, | |
| "eval_loss": 0.9949015378952026, | |
| "eval_runtime": 367.6755, | |
| "eval_samples_per_second": 172.552, | |
| "eval_steps_per_second": 5.393, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 4.542606516290727, | |
| "grad_norm": 2.6259942054748535, | |
| "learning_rate": 5.65e-05, | |
| "loss": 0.9685, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.542606516290727, | |
| "eval_accuracy": 0.8042554828282609, | |
| "eval_loss": 0.9936387538909912, | |
| "eval_runtime": 368.1053, | |
| "eval_samples_per_second": 172.35, | |
| "eval_steps_per_second": 5.387, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 4.594820384294069, | |
| "grad_norm": 3.246826648712158, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.9736, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.594820384294069, | |
| "eval_accuracy": 0.8043477668789687, | |
| "eval_loss": 0.9914335608482361, | |
| "eval_runtime": 367.7482, | |
| "eval_samples_per_second": 172.518, | |
| "eval_steps_per_second": 5.392, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 4.64703425229741, | |
| "grad_norm": 2.689743995666504, | |
| "learning_rate": 5.550000000000001e-05, | |
| "loss": 0.9695, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.64703425229741, | |
| "eval_accuracy": 0.8044957424284033, | |
| "eval_loss": 1.0012022256851196, | |
| "eval_runtime": 367.8322, | |
| "eval_samples_per_second": 172.478, | |
| "eval_steps_per_second": 5.391, | |
| "step": 44500 | |
| }, | |
| { | |
| "epoch": 4.6992481203007515, | |
| "grad_norm": 2.6002440452575684, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.9632, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.6992481203007515, | |
| "eval_accuracy": 0.8053408897805091, | |
| "eval_loss": 0.9931978583335876, | |
| "eval_runtime": 367.8417, | |
| "eval_samples_per_second": 172.474, | |
| "eval_steps_per_second": 5.391, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 4.751461988304094, | |
| "grad_norm": 2.720722198486328, | |
| "learning_rate": 5.45e-05, | |
| "loss": 0.9683, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.751461988304094, | |
| "eval_accuracy": 0.8055148927829874, | |
| "eval_loss": 0.9913645386695862, | |
| "eval_runtime": 367.8141, | |
| "eval_samples_per_second": 172.487, | |
| "eval_steps_per_second": 5.391, | |
| "step": 45500 | |
| }, | |
| { | |
| "epoch": 4.803675856307435, | |
| "grad_norm": 2.6984803676605225, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 0.9559, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.803675856307435, | |
| "eval_accuracy": 0.8059522006843057, | |
| "eval_loss": 0.9891652464866638, | |
| "eval_runtime": 368.2454, | |
| "eval_samples_per_second": 172.285, | |
| "eval_steps_per_second": 5.385, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 4.855889724310777, | |
| "grad_norm": 2.4419870376586914, | |
| "learning_rate": 5.3500000000000006e-05, | |
| "loss": 0.9592, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.855889724310777, | |
| "eval_accuracy": 0.805452993603716, | |
| "eval_loss": 0.9957113862037659, | |
| "eval_runtime": 368.2591, | |
| "eval_samples_per_second": 172.278, | |
| "eval_steps_per_second": 5.385, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 4.908103592314118, | |
| "grad_norm": 2.915653944015503, | |
| "learning_rate": 5.300000000000001e-05, | |
| "loss": 0.9593, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.908103592314118, | |
| "eval_accuracy": 0.8068829182930286, | |
| "eval_loss": 0.987449049949646, | |
| "eval_runtime": 367.9346, | |
| "eval_samples_per_second": 172.43, | |
| "eval_steps_per_second": 5.39, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 4.9603174603174605, | |
| "grad_norm": 2.3721494674682617, | |
| "learning_rate": 5.25e-05, | |
| "loss": 0.9521, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 4.9603174603174605, | |
| "eval_accuracy": 0.8072047578358524, | |
| "eval_loss": 0.9854400157928467, | |
| "eval_runtime": 368.1441, | |
| "eval_samples_per_second": 172.332, | |
| "eval_steps_per_second": 5.386, | |
| "step": 47500 | |
| }, | |
| { | |
| "epoch": 5.012531328320802, | |
| "grad_norm": 2.7081363201141357, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 0.9516, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.012531328320802, | |
| "eval_accuracy": 0.8074420340361171, | |
| "eval_loss": 0.9737870097160339, | |
| "eval_runtime": 367.5696, | |
| "eval_samples_per_second": 172.601, | |
| "eval_steps_per_second": 5.395, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 5.064745196324144, | |
| "grad_norm": 2.834805965423584, | |
| "learning_rate": 5.1500000000000005e-05, | |
| "loss": 0.9361, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.064745196324144, | |
| "eval_accuracy": 0.807780889243512, | |
| "eval_loss": NaN, | |
| "eval_runtime": 366.5337, | |
| "eval_samples_per_second": 173.089, | |
| "eval_steps_per_second": 5.41, | |
| "step": 48500 | |
| }, | |
| { | |
| "epoch": 5.116959064327485, | |
| "grad_norm": 2.5488121509552, | |
| "learning_rate": 5.1000000000000006e-05, | |
| "loss": 0.9472, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.116959064327485, | |
| "eval_accuracy": 0.8083244761680142, | |
| "eval_loss": 0.9714484214782715, | |
| "eval_runtime": 364.0778, | |
| "eval_samples_per_second": 174.257, | |
| "eval_steps_per_second": 5.447, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 5.169172932330827, | |
| "grad_norm": 2.5738179683685303, | |
| "learning_rate": 5.05e-05, | |
| "loss": 0.9387, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.169172932330827, | |
| "eval_accuracy": 0.8083237627586395, | |
| "eval_loss": NaN, | |
| "eval_runtime": 366.0302, | |
| "eval_samples_per_second": 173.327, | |
| "eval_steps_per_second": 5.418, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 5.221386800334169, | |
| "grad_norm": 2.578078508377075, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9405, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.221386800334169, | |
| "eval_accuracy": 0.8092523232767513, | |
| "eval_loss": 0.9723265767097473, | |
| "eval_runtime": 365.9623, | |
| "eval_samples_per_second": 173.359, | |
| "eval_steps_per_second": 5.419, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 5.273600668337511, | |
| "grad_norm": 3.3929808139801025, | |
| "learning_rate": 4.9500000000000004e-05, | |
| "loss": 0.9386, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.273600668337511, | |
| "eval_accuracy": 0.8091146550523858, | |
| "eval_loss": 0.9661393761634827, | |
| "eval_runtime": 365.1784, | |
| "eval_samples_per_second": 173.732, | |
| "eval_steps_per_second": 5.43, | |
| "step": 50500 | |
| }, | |
| { | |
| "epoch": 5.325814536340852, | |
| "grad_norm": 2.3634896278381348, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.933, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.325814536340852, | |
| "eval_accuracy": 0.808835748295871, | |
| "eval_loss": 0.9813971519470215, | |
| "eval_runtime": 381.116, | |
| "eval_samples_per_second": 166.466, | |
| "eval_steps_per_second": 5.203, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 5.378028404344194, | |
| "grad_norm": 2.9042844772338867, | |
| "learning_rate": 4.85e-05, | |
| "loss": 0.9295, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 5.378028404344194, | |
| "eval_accuracy": 0.8103665776447174, | |
| "eval_loss": 0.9754624366760254, | |
| "eval_runtime": 396.2254, | |
| "eval_samples_per_second": 160.118, | |
| "eval_steps_per_second": 5.005, | |
| "step": 51500 | |
| }, | |
| { | |
| "epoch": 5.430242272347535, | |
| "grad_norm": 2.7546634674072266, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9367, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.430242272347535, | |
| "eval_accuracy": 0.8098531380075188, | |
| "eval_loss": 0.9821533560752869, | |
| "eval_runtime": 400.7427, | |
| "eval_samples_per_second": 158.314, | |
| "eval_steps_per_second": 4.948, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 5.482456140350878, | |
| "grad_norm": 2.8839492797851562, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.9292, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 5.482456140350878, | |
| "eval_accuracy": 0.8099045387069566, | |
| "eval_loss": 0.974047064781189, | |
| "eval_runtime": 399.7171, | |
| "eval_samples_per_second": 158.72, | |
| "eval_steps_per_second": 4.961, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 5.534670008354219, | |
| "grad_norm": 2.3632848262786865, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.9217, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 5.534670008354219, | |
| "eval_accuracy": 0.8100555138007703, | |
| "eval_loss": 0.9726797938346863, | |
| "eval_runtime": 398.2666, | |
| "eval_samples_per_second": 159.298, | |
| "eval_steps_per_second": 4.979, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 5.586883876357561, | |
| "grad_norm": 3.127169609069824, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 0.9294, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 5.586883876357561, | |
| "eval_accuracy": 0.8108817419721464, | |
| "eval_loss": 0.9585431218147278, | |
| "eval_runtime": 399.6498, | |
| "eval_samples_per_second": 158.746, | |
| "eval_steps_per_second": 4.962, | |
| "step": 53500 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "grad_norm": 2.53694748878479, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.9195, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 5.639097744360902, | |
| "eval_accuracy": 0.8119484709549883, | |
| "eval_loss": 0.957590639591217, | |
| "eval_runtime": 394.1783, | |
| "eval_samples_per_second": 160.95, | |
| "eval_steps_per_second": 5.031, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 5.6913116123642435, | |
| "grad_norm": 2.6188199520111084, | |
| "learning_rate": 4.55e-05, | |
| "loss": 0.9233, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 5.6913116123642435, | |
| "eval_accuracy": 0.8120400051070349, | |
| "eval_loss": 0.9545953273773193, | |
| "eval_runtime": 395.9182, | |
| "eval_samples_per_second": 160.243, | |
| "eval_steps_per_second": 5.009, | |
| "step": 54500 | |
| }, | |
| { | |
| "epoch": 5.743525480367586, | |
| "grad_norm": 2.502518653869629, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.9272, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 5.743525480367586, | |
| "eval_accuracy": 0.8122659898359591, | |
| "eval_loss": 0.95594322681427, | |
| "eval_runtime": 396.2822, | |
| "eval_samples_per_second": 160.096, | |
| "eval_steps_per_second": 5.004, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 5.795739348370927, | |
| "grad_norm": 2.527303457260132, | |
| "learning_rate": 4.4500000000000004e-05, | |
| "loss": 0.9163, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 5.795739348370927, | |
| "eval_accuracy": 0.8124216614631093, | |
| "eval_loss": 0.9586062431335449, | |
| "eval_runtime": 402.0755, | |
| "eval_samples_per_second": 157.789, | |
| "eval_steps_per_second": 4.932, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 5.847953216374269, | |
| "grad_norm": 2.5056955814361572, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.9249, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.847953216374269, | |
| "eval_accuracy": 0.8123958167743631, | |
| "eval_loss": 0.9501298069953918, | |
| "eval_runtime": 397.3012, | |
| "eval_samples_per_second": 159.685, | |
| "eval_steps_per_second": 4.991, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 5.90016708437761, | |
| "grad_norm": 2.876765727996826, | |
| "learning_rate": 4.35e-05, | |
| "loss": 0.9167, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.90016708437761, | |
| "eval_accuracy": 0.8123634219518265, | |
| "eval_loss": 0.958740770816803, | |
| "eval_runtime": 393.8274, | |
| "eval_samples_per_second": 161.093, | |
| "eval_steps_per_second": 5.035, | |
| "step": 56500 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "grad_norm": 2.457308769226074, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.9191, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 5.9523809523809526, | |
| "eval_accuracy": 0.8134356064347583, | |
| "eval_loss": 0.9526042938232422, | |
| "eval_runtime": 393.3254, | |
| "eval_samples_per_second": 161.299, | |
| "eval_steps_per_second": 5.042, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 6.004594820384294, | |
| "grad_norm": 2.489978075027466, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.9176, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.004594820384294, | |
| "eval_accuracy": 0.8137190312134606, | |
| "eval_loss": 0.9511122703552246, | |
| "eval_runtime": 398.7282, | |
| "eval_samples_per_second": 159.113, | |
| "eval_steps_per_second": 4.973, | |
| "step": 57500 | |
| }, | |
| { | |
| "epoch": 6.056808688387636, | |
| "grad_norm": 2.7550694942474365, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.9034, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.056808688387636, | |
| "eval_accuracy": 0.8137725525224172, | |
| "eval_loss": 0.9519580006599426, | |
| "eval_runtime": 397.3745, | |
| "eval_samples_per_second": 159.655, | |
| "eval_steps_per_second": 4.99, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 6.109022556390977, | |
| "grad_norm": 2.820486068725586, | |
| "learning_rate": 4.15e-05, | |
| "loss": 0.905, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.109022556390977, | |
| "eval_accuracy": 0.8139618722177003, | |
| "eval_loss": 0.9513248801231384, | |
| "eval_runtime": 401.2273, | |
| "eval_samples_per_second": 158.122, | |
| "eval_steps_per_second": 4.942, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 6.161236424394319, | |
| "grad_norm": 2.634119749069214, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.902, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.161236424394319, | |
| "eval_accuracy": 0.8135963134645056, | |
| "eval_loss": 0.9465650916099548, | |
| "eval_runtime": 402.0909, | |
| "eval_samples_per_second": 157.783, | |
| "eval_steps_per_second": 4.932, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 6.213450292397661, | |
| "grad_norm": 2.8152124881744385, | |
| "learning_rate": 4.05e-05, | |
| "loss": 0.9016, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.213450292397661, | |
| "eval_accuracy": 0.8144102857826985, | |
| "eval_loss": 0.9431130886077881, | |
| "eval_runtime": 399.633, | |
| "eval_samples_per_second": 158.753, | |
| "eval_steps_per_second": 4.962, | |
| "step": 59500 | |
| }, | |
| { | |
| "epoch": 6.265664160401003, | |
| "grad_norm": 2.478999137878418, | |
| "learning_rate": 4e-05, | |
| "loss": 0.9018, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.265664160401003, | |
| "eval_accuracy": 0.8148281916671716, | |
| "eval_loss": 0.9465348124504089, | |
| "eval_runtime": 401.047, | |
| "eval_samples_per_second": 158.193, | |
| "eval_steps_per_second": 4.945, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 6.317878028404344, | |
| "grad_norm": 2.4472334384918213, | |
| "learning_rate": 3.9500000000000005e-05, | |
| "loss": 0.8894, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 6.317878028404344, | |
| "eval_accuracy": 0.8152884197022962, | |
| "eval_loss": 0.9336789846420288, | |
| "eval_runtime": 399.7265, | |
| "eval_samples_per_second": 158.716, | |
| "eval_steps_per_second": 4.961, | |
| "step": 60500 | |
| }, | |
| { | |
| "epoch": 6.370091896407686, | |
| "grad_norm": 2.6032795906066895, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 0.8861, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 6.370091896407686, | |
| "eval_accuracy": 0.8158878972595904, | |
| "eval_loss": 0.9414376616477966, | |
| "eval_runtime": 403.8156, | |
| "eval_samples_per_second": 157.109, | |
| "eval_steps_per_second": 4.911, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 6.4223057644110275, | |
| "grad_norm": 2.5782527923583984, | |
| "learning_rate": 3.85e-05, | |
| "loss": 0.8929, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 6.4223057644110275, | |
| "eval_accuracy": 0.8158353407843104, | |
| "eval_loss": 0.9369956254959106, | |
| "eval_runtime": 403.5486, | |
| "eval_samples_per_second": 157.213, | |
| "eval_steps_per_second": 4.914, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 6.474519632414369, | |
| "grad_norm": 2.3801369667053223, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.8869, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 6.474519632414369, | |
| "eval_accuracy": 0.8156502533666622, | |
| "eval_loss": 0.9334683418273926, | |
| "eval_runtime": 401.7901, | |
| "eval_samples_per_second": 157.901, | |
| "eval_steps_per_second": 4.935, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 6.526733500417711, | |
| "grad_norm": 2.44163179397583, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.8933, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 6.526733500417711, | |
| "eval_accuracy": 0.8168851117616432, | |
| "eval_loss": 0.9306558966636658, | |
| "eval_runtime": 397.9786, | |
| "eval_samples_per_second": 159.413, | |
| "eval_steps_per_second": 4.983, | |
| "step": 62500 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "grad_norm": 2.2924203872680664, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.8862, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "eval_accuracy": 0.8162968497056878, | |
| "eval_loss": 0.9353703856468201, | |
| "eval_runtime": 394.7488, | |
| "eval_samples_per_second": 160.717, | |
| "eval_steps_per_second": 5.023, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 6.631161236424394, | |
| "grad_norm": 2.552828550338745, | |
| "learning_rate": 3.65e-05, | |
| "loss": 0.8948, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 6.631161236424394, | |
| "eval_accuracy": 0.8177748484660532, | |
| "eval_loss": 0.9219902157783508, | |
| "eval_runtime": 393.8065, | |
| "eval_samples_per_second": 161.102, | |
| "eval_steps_per_second": 5.035, | |
| "step": 63500 | |
| }, | |
| { | |
| "epoch": 6.683375104427736, | |
| "grad_norm": 2.423050880432129, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.8889, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 6.683375104427736, | |
| "eval_accuracy": 0.8173146779439671, | |
| "eval_loss": 0.9342746138572693, | |
| "eval_runtime": 393.0912, | |
| "eval_samples_per_second": 161.395, | |
| "eval_steps_per_second": 5.045, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 6.735588972431078, | |
| "grad_norm": 2.796268939971924, | |
| "learning_rate": 3.55e-05, | |
| "loss": 0.8885, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 6.735588972431078, | |
| "eval_accuracy": 0.817551689369001, | |
| "eval_loss": 0.9292306303977966, | |
| "eval_runtime": 394.1087, | |
| "eval_samples_per_second": 160.978, | |
| "eval_steps_per_second": 5.032, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 6.787802840434419, | |
| "grad_norm": 2.8128762245178223, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.8854, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 6.787802840434419, | |
| "eval_accuracy": 0.8176226440260885, | |
| "eval_loss": 0.9265376925468445, | |
| "eval_runtime": 394.2997, | |
| "eval_samples_per_second": 160.9, | |
| "eval_steps_per_second": 5.029, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 6.840016708437761, | |
| "grad_norm": 2.8697590827941895, | |
| "learning_rate": 3.45e-05, | |
| "loss": 0.88, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 6.840016708437761, | |
| "eval_accuracy": 0.8185052265994757, | |
| "eval_loss": 0.928463876247406, | |
| "eval_runtime": 394.323, | |
| "eval_samples_per_second": 160.891, | |
| "eval_steps_per_second": 5.029, | |
| "step": 65500 | |
| }, | |
| { | |
| "epoch": 6.8922305764411025, | |
| "grad_norm": 2.6351752281188965, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 0.8841, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.8922305764411025, | |
| "eval_accuracy": 0.8183555493481675, | |
| "eval_loss": 0.9194802045822144, | |
| "eval_runtime": 400.5152, | |
| "eval_samples_per_second": 158.403, | |
| "eval_steps_per_second": 4.951, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 2.2477810382843018, | |
| "learning_rate": 3.35e-05, | |
| "loss": 0.8807, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "eval_accuracy": 0.8187747090050445, | |
| "eval_loss": 0.9138378500938416, | |
| "eval_runtime": 394.4731, | |
| "eval_samples_per_second": 160.83, | |
| "eval_steps_per_second": 5.027, | |
| "step": 66500 | |
| }, | |
| { | |
| "epoch": 6.996658312447786, | |
| "grad_norm": 2.1748201847076416, | |
| "learning_rate": 3.3e-05, | |
| "loss": 0.8836, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 6.996658312447786, | |
| "eval_accuracy": 0.8190003355856446, | |
| "eval_loss": 0.9168543815612793, | |
| "eval_runtime": 394.6502, | |
| "eval_samples_per_second": 160.758, | |
| "eval_steps_per_second": 5.025, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 7.048872180451128, | |
| "grad_norm": 2.7435665130615234, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.8707, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.048872180451128, | |
| "eval_accuracy": 0.8191917867246192, | |
| "eval_loss": 0.9231117367744446, | |
| "eval_runtime": 392.3745, | |
| "eval_samples_per_second": 161.69, | |
| "eval_steps_per_second": 5.054, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 7.101086048454469, | |
| "grad_norm": 2.569931745529175, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8701, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.101086048454469, | |
| "eval_accuracy": 0.819969121038317, | |
| "eval_loss": 0.9153112173080444, | |
| "eval_runtime": 393.0875, | |
| "eval_samples_per_second": 161.397, | |
| "eval_steps_per_second": 5.045, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 7.1532999164578115, | |
| "grad_norm": 2.819972515106201, | |
| "learning_rate": 3.15e-05, | |
| "loss": 0.8656, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 7.1532999164578115, | |
| "eval_accuracy": 0.8194268014568229, | |
| "eval_loss": 0.9146177768707275, | |
| "eval_runtime": 394.2167, | |
| "eval_samples_per_second": 160.934, | |
| "eval_steps_per_second": 5.03, | |
| "step": 68500 | |
| }, | |
| { | |
| "epoch": 7.205513784461153, | |
| "grad_norm": 2.415208578109741, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.871, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 7.205513784461153, | |
| "eval_accuracy": 0.8198532803574815, | |
| "eval_loss": NaN, | |
| "eval_runtime": 395.7273, | |
| "eval_samples_per_second": 160.32, | |
| "eval_steps_per_second": 5.011, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 7.257727652464495, | |
| "grad_norm": 2.7900257110595703, | |
| "learning_rate": 3.05e-05, | |
| "loss": 0.8685, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 7.257727652464495, | |
| "eval_accuracy": 0.8207157152692325, | |
| "eval_loss": 0.9127238392829895, | |
| "eval_runtime": 392.7333, | |
| "eval_samples_per_second": 161.542, | |
| "eval_steps_per_second": 5.049, | |
| "step": 69500 | |
| }, | |
| { | |
| "epoch": 7.309941520467836, | |
| "grad_norm": 2.643833637237549, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8623, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 7.309941520467836, | |
| "eval_accuracy": 0.8202351749773661, | |
| "eval_loss": 0.9171856045722961, | |
| "eval_runtime": 393.097, | |
| "eval_samples_per_second": 161.393, | |
| "eval_steps_per_second": 5.045, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 7.362155388471178, | |
| "grad_norm": 2.5037946701049805, | |
| "learning_rate": 2.95e-05, | |
| "loss": 0.8601, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 7.362155388471178, | |
| "eval_accuracy": 0.8211525544791177, | |
| "eval_loss": 0.903904139995575, | |
| "eval_runtime": 394.3466, | |
| "eval_samples_per_second": 160.881, | |
| "eval_steps_per_second": 5.029, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 7.41436925647452, | |
| "grad_norm": 2.7093377113342285, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.8573, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 7.41436925647452, | |
| "eval_accuracy": 0.8214304681574243, | |
| "eval_loss": 0.9036524891853333, | |
| "eval_runtime": 393.5778, | |
| "eval_samples_per_second": 161.196, | |
| "eval_steps_per_second": 5.038, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 7.466583124477861, | |
| "grad_norm": 2.760674476623535, | |
| "learning_rate": 2.8499999999999998e-05, | |
| "loss": 0.8633, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 7.466583124477861, | |
| "eval_accuracy": 0.8217283854693812, | |
| "eval_loss": 0.9000177979469299, | |
| "eval_runtime": 396.5628, | |
| "eval_samples_per_second": 159.982, | |
| "eval_steps_per_second": 5.0, | |
| "step": 71500 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "grad_norm": 2.831023693084717, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.8576, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 7.518796992481203, | |
| "eval_accuracy": 0.8211579127032714, | |
| "eval_loss": 0.918953001499176, | |
| "eval_runtime": 400.5863, | |
| "eval_samples_per_second": 158.375, | |
| "eval_steps_per_second": 4.95, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 7.571010860484545, | |
| "grad_norm": 2.453390598297119, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.854, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 7.571010860484545, | |
| "eval_accuracy": 0.8220418032437955, | |
| "eval_loss": 0.9041558504104614, | |
| "eval_runtime": 399.249, | |
| "eval_samples_per_second": 158.906, | |
| "eval_steps_per_second": 4.967, | |
| "step": 72500 | |
| }, | |
| { | |
| "epoch": 7.6232247284878865, | |
| "grad_norm": 2.6029062271118164, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.8615, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 7.6232247284878865, | |
| "eval_accuracy": 0.8223261586980396, | |
| "eval_loss": 0.8994259834289551, | |
| "eval_runtime": 387.0421, | |
| "eval_samples_per_second": 163.918, | |
| "eval_steps_per_second": 5.123, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 7.675438596491228, | |
| "grad_norm": 2.3841817378997803, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 0.8632, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 7.675438596491228, | |
| "eval_accuracy": 0.8218723618637837, | |
| "eval_loss": 0.9080346822738647, | |
| "eval_runtime": 368.6926, | |
| "eval_samples_per_second": 172.076, | |
| "eval_steps_per_second": 5.378, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 7.72765246449457, | |
| "grad_norm": 3.8187203407287598, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.8519, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 7.72765246449457, | |
| "eval_accuracy": 0.8223824804514319, | |
| "eval_loss": 0.9071189761161804, | |
| "eval_runtime": 367.458, | |
| "eval_samples_per_second": 172.654, | |
| "eval_steps_per_second": 5.397, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 7.779866332497911, | |
| "grad_norm": 2.744349241256714, | |
| "learning_rate": 2.5500000000000003e-05, | |
| "loss": 0.8569, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 7.779866332497911, | |
| "eval_accuracy": 0.8229439133116854, | |
| "eval_loss": 0.9003260731697083, | |
| "eval_runtime": 369.188, | |
| "eval_samples_per_second": 171.845, | |
| "eval_steps_per_second": 5.371, | |
| "step": 74500 | |
| }, | |
| { | |
| "epoch": 7.832080200501253, | |
| "grad_norm": 2.3172342777252197, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.8502, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 7.832080200501253, | |
| "eval_accuracy": 0.8229715307093431, | |
| "eval_loss": 0.9045436382293701, | |
| "eval_runtime": 368.1168, | |
| "eval_samples_per_second": 172.345, | |
| "eval_steps_per_second": 5.387, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 7.884294068504595, | |
| "grad_norm": 2.6108663082122803, | |
| "learning_rate": 2.45e-05, | |
| "loss": 0.8536, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 7.884294068504595, | |
| "eval_accuracy": 0.8234089591247552, | |
| "eval_loss": 0.9026762247085571, | |
| "eval_runtime": 366.8391, | |
| "eval_samples_per_second": 172.945, | |
| "eval_steps_per_second": 5.406, | |
| "step": 75500 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "grad_norm": 2.315197467803955, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.85, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 7.936507936507937, | |
| "eval_accuracy": 0.8232359994242249, | |
| "eval_loss": 0.8995440006256104, | |
| "eval_runtime": 367.8582, | |
| "eval_samples_per_second": 172.466, | |
| "eval_steps_per_second": 5.391, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 7.988721804511278, | |
| "grad_norm": 2.611060380935669, | |
| "learning_rate": 2.35e-05, | |
| "loss": 0.8453, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 7.988721804511278, | |
| "eval_accuracy": 0.8234605237364002, | |
| "eval_loss": 0.8912914395332336, | |
| "eval_runtime": 367.6079, | |
| "eval_samples_per_second": 172.583, | |
| "eval_steps_per_second": 5.394, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 8.04093567251462, | |
| "grad_norm": 2.4672865867614746, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.8294, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 8.04093567251462, | |
| "eval_accuracy": 0.8244379262655704, | |
| "eval_loss": 0.896998405456543, | |
| "eval_runtime": 367.8054, | |
| "eval_samples_per_second": 172.491, | |
| "eval_steps_per_second": 5.391, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 8.093149540517961, | |
| "grad_norm": 2.645519495010376, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.849, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 8.093149540517961, | |
| "eval_accuracy": 0.8244201499571945, | |
| "eval_loss": 0.8957926034927368, | |
| "eval_runtime": 369.9508, | |
| "eval_samples_per_second": 171.49, | |
| "eval_steps_per_second": 5.36, | |
| "step": 77500 | |
| }, | |
| { | |
| "epoch": 8.145363408521304, | |
| "grad_norm": 2.7964978218078613, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.8363, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 8.145363408521304, | |
| "eval_accuracy": 0.8244232177017738, | |
| "eval_loss": 0.8884155750274658, | |
| "eval_runtime": 367.358, | |
| "eval_samples_per_second": 172.701, | |
| "eval_steps_per_second": 5.398, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 8.197577276524646, | |
| "grad_norm": 2.48044490814209, | |
| "learning_rate": 2.15e-05, | |
| "loss": 0.8355, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 8.197577276524646, | |
| "eval_accuracy": 0.82532828384256, | |
| "eval_loss": 0.8786827325820923, | |
| "eval_runtime": 366.707, | |
| "eval_samples_per_second": 173.007, | |
| "eval_steps_per_second": 5.408, | |
| "step": 78500 | |
| }, | |
| { | |
| "epoch": 8.249791144527986, | |
| "grad_norm": 2.6931161880493164, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.8439, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 8.249791144527986, | |
| "eval_accuracy": 0.8246382720646043, | |
| "eval_loss": 0.8848527073860168, | |
| "eval_runtime": 366.9894, | |
| "eval_samples_per_second": 172.874, | |
| "eval_steps_per_second": 5.403, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 8.302005012531328, | |
| "grad_norm": 2.607593297958374, | |
| "learning_rate": 2.05e-05, | |
| "loss": 0.8378, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 8.302005012531328, | |
| "eval_accuracy": 0.8252490751457086, | |
| "eval_loss": NaN, | |
| "eval_runtime": 366.5162, | |
| "eval_samples_per_second": 173.097, | |
| "eval_steps_per_second": 5.41, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 8.35421888053467, | |
| "grad_norm": 2.4364349842071533, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8322, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 8.35421888053467, | |
| "eval_accuracy": 0.8252812537780041, | |
| "eval_loss": NaN, | |
| "eval_runtime": 367.8653, | |
| "eval_samples_per_second": 172.463, | |
| "eval_steps_per_second": 5.391, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 8.406432748538013, | |
| "grad_norm": 2.541379928588867, | |
| "learning_rate": 1.9500000000000003e-05, | |
| "loss": 0.8366, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 8.406432748538013, | |
| "eval_accuracy": 0.8252309925385912, | |
| "eval_loss": 0.9011866450309753, | |
| "eval_runtime": 367.843, | |
| "eval_samples_per_second": 172.473, | |
| "eval_steps_per_second": 5.391, | |
| "step": 80500 | |
| }, | |
| { | |
| "epoch": 8.458646616541353, | |
| "grad_norm": 2.3451640605926514, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.8313, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 8.458646616541353, | |
| "eval_accuracy": 0.8259800217048817, | |
| "eval_loss": 0.8837263584136963, | |
| "eval_runtime": 367.6187, | |
| "eval_samples_per_second": 172.578, | |
| "eval_steps_per_second": 5.394, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 8.510860484544695, | |
| "grad_norm": 2.6211562156677246, | |
| "learning_rate": 1.85e-05, | |
| "loss": 0.8289, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 8.510860484544695, | |
| "eval_accuracy": 0.8257132907861338, | |
| "eval_loss": 0.8861507773399353, | |
| "eval_runtime": 367.3072, | |
| "eval_samples_per_second": 172.725, | |
| "eval_steps_per_second": 5.399, | |
| "step": 81500 | |
| }, | |
| { | |
| "epoch": 8.563074352548037, | |
| "grad_norm": 2.1193125247955322, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.8337, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 8.563074352548037, | |
| "eval_accuracy": 0.8263971078151469, | |
| "eval_loss": 0.8711854815483093, | |
| "eval_runtime": 366.8982, | |
| "eval_samples_per_second": 172.917, | |
| "eval_steps_per_second": 5.405, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 8.615288220551378, | |
| "grad_norm": 2.5682766437530518, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.8207, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 8.615288220551378, | |
| "eval_accuracy": 0.8263459457288101, | |
| "eval_loss": 0.8814241886138916, | |
| "eval_runtime": 366.9346, | |
| "eval_samples_per_second": 172.9, | |
| "eval_steps_per_second": 5.404, | |
| "step": 82500 | |
| }, | |
| { | |
| "epoch": 8.66750208855472, | |
| "grad_norm": 2.460284471511841, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.8273, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 8.66750208855472, | |
| "eval_accuracy": 0.8265486834185438, | |
| "eval_loss": 0.8726556301116943, | |
| "eval_runtime": 367.479, | |
| "eval_samples_per_second": 172.644, | |
| "eval_steps_per_second": 5.396, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 8.719715956558062, | |
| "grad_norm": 2.519878625869751, | |
| "learning_rate": 1.65e-05, | |
| "loss": 0.8212, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 8.719715956558062, | |
| "eval_accuracy": 0.8265830712467154, | |
| "eval_loss": 0.8803038001060486, | |
| "eval_runtime": 368.0421, | |
| "eval_samples_per_second": 172.38, | |
| "eval_steps_per_second": 5.388, | |
| "step": 83500 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 2.9779090881347656, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.8267, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "eval_accuracy": 0.8266352426703358, | |
| "eval_loss": NaN, | |
| "eval_runtime": 370.1684, | |
| "eval_samples_per_second": 171.39, | |
| "eval_steps_per_second": 5.357, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 8.824143692564745, | |
| "grad_norm": 2.5584726333618164, | |
| "learning_rate": 1.55e-05, | |
| "loss": 0.8236, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 8.824143692564745, | |
| "eval_accuracy": 0.8272979799699072, | |
| "eval_loss": 0.8709605932235718, | |
| "eval_runtime": 367.2687, | |
| "eval_samples_per_second": 172.743, | |
| "eval_steps_per_second": 5.399, | |
| "step": 84500 | |
| }, | |
| { | |
| "epoch": 8.876357560568087, | |
| "grad_norm": 2.8752248287200928, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.8284, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 8.876357560568087, | |
| "eval_accuracy": 0.8270593176674245, | |
| "eval_loss": 0.8703105449676514, | |
| "eval_runtime": 367.121, | |
| "eval_samples_per_second": 172.812, | |
| "eval_steps_per_second": 5.401, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "grad_norm": 2.441423177719116, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.8305, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "eval_accuracy": 0.8273055974821587, | |
| "eval_loss": 0.8782740831375122, | |
| "eval_runtime": 367.2074, | |
| "eval_samples_per_second": 172.772, | |
| "eval_steps_per_second": 5.4, | |
| "step": 85500 | |
| }, | |
| { | |
| "epoch": 8.980785296574771, | |
| "grad_norm": 2.5627357959747314, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 0.824, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 8.980785296574771, | |
| "eval_accuracy": 0.8278787560948533, | |
| "eval_loss": 0.8767423033714294, | |
| "eval_runtime": 367.8162, | |
| "eval_samples_per_second": 172.486, | |
| "eval_steps_per_second": 5.391, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 9.032999164578111, | |
| "grad_norm": 2.9659039974212646, | |
| "learning_rate": 1.3500000000000001e-05, | |
| "loss": 0.8111, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 9.032999164578111, | |
| "eval_accuracy": 0.8281429418603345, | |
| "eval_loss": NaN, | |
| "eval_runtime": 365.8491, | |
| "eval_samples_per_second": 173.413, | |
| "eval_steps_per_second": 5.42, | |
| "step": 86500 | |
| }, | |
| { | |
| "epoch": 9.085213032581454, | |
| "grad_norm": 2.6677966117858887, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 0.8114, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 9.085213032581454, | |
| "eval_accuracy": 0.8280896401382127, | |
| "eval_loss": 0.8696718215942383, | |
| "eval_runtime": 368.939, | |
| "eval_samples_per_second": 171.961, | |
| "eval_steps_per_second": 5.375, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 9.137426900584796, | |
| "grad_norm": 2.8159141540527344, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.8214, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 9.137426900584796, | |
| "eval_accuracy": 0.8287721660611849, | |
| "eval_loss": 0.865568995475769, | |
| "eval_runtime": 367.5282, | |
| "eval_samples_per_second": 172.621, | |
| "eval_steps_per_second": 5.396, | |
| "step": 87500 | |
| }, | |
| { | |
| "epoch": 9.189640768588138, | |
| "grad_norm": 2.44439959526062, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.8117, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 9.189640768588138, | |
| "eval_accuracy": 0.8286521507089889, | |
| "eval_loss": 0.8658538460731506, | |
| "eval_runtime": 365.9254, | |
| "eval_samples_per_second": 173.377, | |
| "eval_steps_per_second": 5.419, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 9.241854636591478, | |
| "grad_norm": 2.686903953552246, | |
| "learning_rate": 1.1500000000000002e-05, | |
| "loss": 0.8176, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 9.241854636591478, | |
| "eval_accuracy": 0.8286974413686677, | |
| "eval_loss": 0.8724528551101685, | |
| "eval_runtime": 366.9175, | |
| "eval_samples_per_second": 172.908, | |
| "eval_steps_per_second": 5.404, | |
| "step": 88500 | |
| }, | |
| { | |
| "epoch": 9.29406850459482, | |
| "grad_norm": 2.9780402183532715, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.8098, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 9.29406850459482, | |
| "eval_accuracy": 0.8288113105821936, | |
| "eval_loss": 0.8693479895591736, | |
| "eval_runtime": 366.5706, | |
| "eval_samples_per_second": 173.072, | |
| "eval_steps_per_second": 5.41, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 9.346282372598163, | |
| "grad_norm": 2.3616678714752197, | |
| "learning_rate": 1.05e-05, | |
| "loss": 0.8101, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 9.346282372598163, | |
| "eval_accuracy": 0.8288534544077049, | |
| "eval_loss": 0.8676702380180359, | |
| "eval_runtime": 367.6956, | |
| "eval_samples_per_second": 172.542, | |
| "eval_steps_per_second": 5.393, | |
| "step": 89500 | |
| }, | |
| { | |
| "epoch": 9.398496240601503, | |
| "grad_norm": 2.4607014656066895, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8074, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 9.398496240601503, | |
| "eval_accuracy": 0.8293368416792034, | |
| "eval_loss": 0.868602454662323, | |
| "eval_runtime": 370.3428, | |
| "eval_samples_per_second": 171.309, | |
| "eval_steps_per_second": 5.354, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 9.450710108604845, | |
| "grad_norm": 2.303865432739258, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.8075, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 9.450710108604845, | |
| "eval_accuracy": 0.829557819292624, | |
| "eval_loss": 0.8760720491409302, | |
| "eval_runtime": 366.9515, | |
| "eval_samples_per_second": 172.892, | |
| "eval_steps_per_second": 5.404, | |
| "step": 90500 | |
| }, | |
| { | |
| "epoch": 9.502923976608187, | |
| "grad_norm": 3.1800107955932617, | |
| "learning_rate": 9e-06, | |
| "loss": 0.813, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 9.502923976608187, | |
| "eval_accuracy": 0.8293719470428488, | |
| "eval_loss": 0.8725053668022156, | |
| "eval_runtime": 366.7268, | |
| "eval_samples_per_second": 172.998, | |
| "eval_steps_per_second": 5.407, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 9.55513784461153, | |
| "grad_norm": 2.895458698272705, | |
| "learning_rate": 8.500000000000002e-06, | |
| "loss": 0.8142, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 9.55513784461153, | |
| "eval_accuracy": 0.8299493767535929, | |
| "eval_loss": 0.8597660660743713, | |
| "eval_runtime": 366.9591, | |
| "eval_samples_per_second": 172.888, | |
| "eval_steps_per_second": 5.404, | |
| "step": 91500 | |
| }, | |
| { | |
| "epoch": 9.60735171261487, | |
| "grad_norm": 2.494898796081543, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8028, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 9.60735171261487, | |
| "eval_accuracy": 0.8296925250842329, | |
| "eval_loss": 0.8663123250007629, | |
| "eval_runtime": 366.9071, | |
| "eval_samples_per_second": 172.913, | |
| "eval_steps_per_second": 5.405, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 9.659565580618212, | |
| "grad_norm": 2.6649532318115234, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.8051, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 9.659565580618212, | |
| "eval_accuracy": 0.8297183234153672, | |
| "eval_loss": 0.8609718084335327, | |
| "eval_runtime": 369.3381, | |
| "eval_samples_per_second": 171.775, | |
| "eval_steps_per_second": 5.369, | |
| "step": 92500 | |
| }, | |
| { | |
| "epoch": 9.711779448621554, | |
| "grad_norm": 2.708948850631714, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.7982, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 9.711779448621554, | |
| "eval_accuracy": 0.8295886536989987, | |
| "eval_loss": 0.8697434067726135, | |
| "eval_runtime": 367.9632, | |
| "eval_samples_per_second": 172.417, | |
| "eval_steps_per_second": 5.389, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 9.763993316624896, | |
| "grad_norm": 2.320215940475464, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 0.7985, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 9.763993316624896, | |
| "eval_accuracy": 0.8301079266760504, | |
| "eval_loss": 0.8655940890312195, | |
| "eval_runtime": 366.2895, | |
| "eval_samples_per_second": 173.205, | |
| "eval_steps_per_second": 5.414, | |
| "step": 93500 | |
| }, | |
| { | |
| "epoch": 9.816207184628237, | |
| "grad_norm": 2.776233673095703, | |
| "learning_rate": 6e-06, | |
| "loss": 0.8016, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 9.816207184628237, | |
| "eval_accuracy": 0.8302324108540118, | |
| "eval_loss": 0.8625151515007019, | |
| "eval_runtime": 367.8541, | |
| "eval_samples_per_second": 172.468, | |
| "eval_steps_per_second": 5.391, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 9.868421052631579, | |
| "grad_norm": 2.461749792098999, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 0.8044, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 9.868421052631579, | |
| "eval_accuracy": 0.8303814560823992, | |
| "eval_loss": NaN, | |
| "eval_runtime": 366.9598, | |
| "eval_samples_per_second": 172.888, | |
| "eval_steps_per_second": 5.404, | |
| "step": 94500 | |
| }, | |
| { | |
| "epoch": 9.920634920634921, | |
| "grad_norm": 2.611931085586548, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 9.920634920634921, | |
| "eval_accuracy": 0.8306366115947627, | |
| "eval_loss": 0.8623033165931702, | |
| "eval_runtime": 368.1962, | |
| "eval_samples_per_second": 172.308, | |
| "eval_steps_per_second": 5.386, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 9.972848788638263, | |
| "grad_norm": 2.500162363052368, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.7972, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 9.972848788638263, | |
| "eval_accuracy": 0.8306005889087491, | |
| "eval_loss": 0.8600591421127319, | |
| "eval_runtime": 369.0733, | |
| "eval_samples_per_second": 171.898, | |
| "eval_steps_per_second": 5.373, | |
| "step": 95500 | |
| }, | |
| { | |
| "epoch": 10.025062656641603, | |
| "grad_norm": 3.495044231414795, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.7929, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 10.025062656641603, | |
| "eval_accuracy": 0.8308683319034997, | |
| "eval_loss": 0.8547109961509705, | |
| "eval_runtime": 368.4222, | |
| "eval_samples_per_second": 172.202, | |
| "eval_steps_per_second": 5.382, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 10.077276524644946, | |
| "grad_norm": 2.7070469856262207, | |
| "learning_rate": 3.5000000000000004e-06, | |
| "loss": 0.7903, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 10.077276524644946, | |
| "eval_accuracy": 0.8307715292279126, | |
| "eval_loss": 0.8600655198097229, | |
| "eval_runtime": 366.9772, | |
| "eval_samples_per_second": 172.88, | |
| "eval_steps_per_second": 5.404, | |
| "step": 96500 | |
| }, | |
| { | |
| "epoch": 10.129490392648288, | |
| "grad_norm": 3.4624171257019043, | |
| "learning_rate": 3e-06, | |
| "loss": 0.7898, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 10.129490392648288, | |
| "eval_accuracy": 0.831032096060647, | |
| "eval_loss": 0.853934109210968, | |
| "eval_runtime": 368.4043, | |
| "eval_samples_per_second": 172.21, | |
| "eval_steps_per_second": 5.383, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 10.18170426065163, | |
| "grad_norm": 2.6027026176452637, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.7994, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 10.18170426065163, | |
| "eval_accuracy": 0.8308584971832438, | |
| "eval_loss": 0.8530024886131287, | |
| "eval_runtime": 366.5565, | |
| "eval_samples_per_second": 173.078, | |
| "eval_steps_per_second": 5.41, | |
| "step": 97500 | |
| }, | |
| { | |
| "epoch": 10.23391812865497, | |
| "grad_norm": 2.5377743244171143, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.7926, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 10.23391812865497, | |
| "eval_accuracy": 0.8310957564926591, | |
| "eval_loss": 0.8588021993637085, | |
| "eval_runtime": 366.0793, | |
| "eval_samples_per_second": 173.304, | |
| "eval_steps_per_second": 5.417, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 10.286131996658312, | |
| "grad_norm": 2.9179139137268066, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.785, | |
| "step": 98500 | |
| }, | |
| { | |
| "epoch": 10.286131996658312, | |
| "eval_accuracy": 0.8314594660073593, | |
| "eval_loss": 0.8528650403022766, | |
| "eval_runtime": 366.8402, | |
| "eval_samples_per_second": 172.945, | |
| "eval_steps_per_second": 5.406, | |
| "step": 98500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 100000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "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": 8.304812876669911e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |