| { | |
| "best_global_step": 1000, | |
| "best_metric": 9.911575317382812, | |
| "best_model_checkpoint": "./models/v-001/checkpoint-1000", | |
| "epoch": 64.55434782608695, | |
| "eval_steps": 100, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6521739130434783, | |
| "grad_norm": 0.5281125903129578, | |
| "learning_rate": 3.166666666666667e-06, | |
| "loss": 10.3483, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.2934782608695652, | |
| "grad_norm": 0.6133605241775513, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 10.3417, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.9456521739130435, | |
| "grad_norm": 0.6125457882881165, | |
| "learning_rate": 9.833333333333333e-06, | |
| "loss": 10.3299, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.5869565217391304, | |
| "grad_norm": 0.5962333679199219, | |
| "learning_rate": 1.3166666666666665e-05, | |
| "loss": 10.3064, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2282608695652173, | |
| "grad_norm": 0.6132860779762268, | |
| "learning_rate": 1.65e-05, | |
| "loss": 10.2727, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.2282608695652173, | |
| "eval_accuracy": 4.39651795777744e-05, | |
| "eval_loss": 10.328398704528809, | |
| "eval_runtime": 17.8533, | |
| "eval_samples_per_second": 16.916, | |
| "eval_steps_per_second": 1.064, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.880434782608696, | |
| "grad_norm": 0.5671436190605164, | |
| "learning_rate": 1.9833333333333335e-05, | |
| "loss": 10.2083, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.4685352146625519, | |
| "learning_rate": 2.3166666666666666e-05, | |
| "loss": 10.1064, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.163043478260869, | |
| "grad_norm": 0.45952484011650085, | |
| "learning_rate": 2.6500000000000004e-05, | |
| "loss": 9.9909, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.815217391304348, | |
| "grad_norm": 0.41472023725509644, | |
| "learning_rate": 2.9833333333333335e-05, | |
| "loss": 9.8718, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.456521739130435, | |
| "grad_norm": 0.43127089738845825, | |
| "learning_rate": 3.316666666666667e-05, | |
| "loss": 9.7582, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.456521739130435, | |
| "eval_accuracy": 0.0026139297676240417, | |
| "eval_loss": 10.096575736999512, | |
| "eval_runtime": 35.2536, | |
| "eval_samples_per_second": 8.566, | |
| "eval_steps_per_second": 0.539, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.0978260869565215, | |
| "grad_norm": 0.4042549431324005, | |
| "learning_rate": 3.65e-05, | |
| "loss": 9.6266, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 0.3853429853916168, | |
| "learning_rate": 3.983333333333333e-05, | |
| "loss": 9.5189, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.391304347826088, | |
| "grad_norm": 0.38628196716308594, | |
| "learning_rate": 4.316666666666667e-05, | |
| "loss": 9.4044, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.032608695652174, | |
| "grad_norm": 0.43260782957077026, | |
| "learning_rate": 4.6500000000000005e-05, | |
| "loss": 9.3052, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 9.684782608695652, | |
| "grad_norm": 0.4363991916179657, | |
| "learning_rate": 4.9833333333333336e-05, | |
| "loss": 9.2052, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.684782608695652, | |
| "eval_accuracy": 0.003677087746504768, | |
| "eval_loss": 9.951318740844727, | |
| "eval_runtime": 47.0129, | |
| "eval_samples_per_second": 6.424, | |
| "eval_steps_per_second": 0.404, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 10.326086956521738, | |
| "grad_norm": 0.49219855666160583, | |
| "learning_rate": 5.316666666666667e-05, | |
| "loss": 9.1327, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 10.978260869565217, | |
| "grad_norm": 0.4471158981323242, | |
| "learning_rate": 5.65e-05, | |
| "loss": 9.0328, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 11.619565217391305, | |
| "grad_norm": 0.3913232684135437, | |
| "learning_rate": 5.983333333333334e-05, | |
| "loss": 8.9514, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 12.26086956521739, | |
| "grad_norm": 1.0882291793823242, | |
| "learning_rate": 6.316666666666668e-05, | |
| "loss": 8.8916, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 12.91304347826087, | |
| "grad_norm": 0.6010486483573914, | |
| "learning_rate": 6.65e-05, | |
| "loss": 8.8216, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 12.91304347826087, | |
| "eval_accuracy": 0.003413296669038122, | |
| "eval_loss": 9.953831672668457, | |
| "eval_runtime": 36.8026, | |
| "eval_samples_per_second": 8.206, | |
| "eval_steps_per_second": 0.516, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 13.554347826086957, | |
| "grad_norm": 0.49556687474250793, | |
| "learning_rate": 6.983333333333334e-05, | |
| "loss": 8.7406, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 14.195652173913043, | |
| "grad_norm": 0.495381623506546, | |
| "learning_rate": 7.316666666666668e-05, | |
| "loss": 8.679, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 14.847826086956522, | |
| "grad_norm": 0.6165482401847839, | |
| "learning_rate": 7.65e-05, | |
| "loss": 8.5904, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 15.48913043478261, | |
| "grad_norm": 0.5654007792472839, | |
| "learning_rate": 7.983333333333334e-05, | |
| "loss": 8.4896, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 16.130434782608695, | |
| "grad_norm": 0.6611935496330261, | |
| "learning_rate": 8.316666666666666e-05, | |
| "loss": 8.406, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.130434782608695, | |
| "eval_accuracy": 0.0029496638662179554, | |
| "eval_loss": 9.952414512634277, | |
| "eval_runtime": 39.649, | |
| "eval_samples_per_second": 7.617, | |
| "eval_steps_per_second": 0.479, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 16.782608695652176, | |
| "grad_norm": 0.7537912726402283, | |
| "learning_rate": 8.65e-05, | |
| "loss": 8.2948, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 17.42391304347826, | |
| "grad_norm": 0.9145230650901794, | |
| "learning_rate": 8.983333333333334e-05, | |
| "loss": 8.1992, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 18.065217391304348, | |
| "grad_norm": 0.810655415058136, | |
| "learning_rate": 9.316666666666666e-05, | |
| "loss": 8.08, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 18.717391304347824, | |
| "grad_norm": 0.9121057987213135, | |
| "learning_rate": 9.65e-05, | |
| "loss": 7.9438, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 19.358695652173914, | |
| "grad_norm": 0.8612993359565735, | |
| "learning_rate": 9.983333333333334e-05, | |
| "loss": 7.8326, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 19.358695652173914, | |
| "eval_accuracy": 0.0021423032957897346, | |
| "eval_loss": 9.945837020874023, | |
| "eval_runtime": 25.9715, | |
| "eval_samples_per_second": 11.628, | |
| "eval_steps_per_second": 0.732, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.8960981369018555, | |
| "learning_rate": 9.995456138403733e-05, | |
| "loss": 7.716, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 20.652173913043477, | |
| "grad_norm": 0.9986662268638611, | |
| "learning_rate": 9.980864681729001e-05, | |
| "loss": 7.5692, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 21.293478260869566, | |
| "grad_norm": 0.9232766628265381, | |
| "learning_rate": 9.956242426451834e-05, | |
| "loss": 7.4208, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 21.945652173913043, | |
| "grad_norm": 0.9031963348388672, | |
| "learning_rate": 9.921638958517565e-05, | |
| "loss": 7.3481, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 22.58695652173913, | |
| "grad_norm": 0.9567400813102722, | |
| "learning_rate": 9.877123964705497e-05, | |
| "loss": 7.1956, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 22.58695652173913, | |
| "eval_accuracy": 0.001666679989448357, | |
| "eval_loss": 9.986405372619629, | |
| "eval_runtime": 57.7614, | |
| "eval_samples_per_second": 5.228, | |
| "eval_steps_per_second": 0.329, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 23.22826086956522, | |
| "grad_norm": 1.135650873184204, | |
| "learning_rate": 9.822787092288991e-05, | |
| "loss": 7.0604, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 23.880434782608695, | |
| "grad_norm": 0.8771520853042603, | |
| "learning_rate": 9.758737768497802e-05, | |
| "loss": 6.9215, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 24.52173913043478, | |
| "grad_norm": 1.0156564712524414, | |
| "learning_rate": 9.685104980146193e-05, | |
| "loss": 6.8363, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 25.16304347826087, | |
| "grad_norm": 0.9963734149932861, | |
| "learning_rate": 9.60203701387066e-05, | |
| "loss": 6.6577, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 25.815217391304348, | |
| "grad_norm": 0.8723818063735962, | |
| "learning_rate": 9.509701157500376e-05, | |
| "loss": 6.5659, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 25.815217391304348, | |
| "eval_accuracy": 0.0014988129401514, | |
| "eval_loss": 9.925810813903809, | |
| "eval_runtime": 49.4048, | |
| "eval_samples_per_second": 6.113, | |
| "eval_steps_per_second": 0.385, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 26.456521739130434, | |
| "grad_norm": 1.1702412366867065, | |
| "learning_rate": 9.408283363161774e-05, | |
| "loss": 6.393, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 27.097826086956523, | |
| "grad_norm": 0.8747526407241821, | |
| "learning_rate": 9.297987872795705e-05, | |
| "loss": 6.3074, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 27.75, | |
| "grad_norm": 0.973866879940033, | |
| "learning_rate": 9.179036806841353e-05, | |
| "loss": 6.1801, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 28.391304347826086, | |
| "grad_norm": 1.585481882095337, | |
| "learning_rate": 9.051669716915227e-05, | |
| "loss": 6.1175, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 29.032608695652176, | |
| "grad_norm": 1.0919766426086426, | |
| "learning_rate": 8.916143103386093e-05, | |
| "loss": 5.9719, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 29.032608695652176, | |
| "eval_accuracy": 0.001458844595080696, | |
| "eval_loss": 9.97097396850586, | |
| "eval_runtime": 46.643, | |
| "eval_samples_per_second": 6.475, | |
| "eval_steps_per_second": 0.407, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 29.684782608695652, | |
| "grad_norm": 1.173614501953125, | |
| "learning_rate": 8.77272989881736e-05, | |
| "loss": 5.8702, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 30.32608695652174, | |
| "grad_norm": 1.0350476503372192, | |
| "learning_rate": 8.621718918317225e-05, | |
| "loss": 5.704, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 30.97826086956522, | |
| "grad_norm": 1.1128321886062622, | |
| "learning_rate": 8.463414277903475e-05, | |
| "loss": 5.6413, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 31.619565217391305, | |
| "grad_norm": 1.2460695505142212, | |
| "learning_rate": 8.298134782054305e-05, | |
| "loss": 5.4948, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 32.26086956521739, | |
| "grad_norm": 1.1606298685073853, | |
| "learning_rate": 8.126213281678526e-05, | |
| "loss": 5.4031, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 32.26086956521739, | |
| "eval_accuracy": 0.001079145316909008, | |
| "eval_loss": 9.911575317382812, | |
| "eval_runtime": 48.65, | |
| "eval_samples_per_second": 6.208, | |
| "eval_steps_per_second": 0.391, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 32.91304347826087, | |
| "grad_norm": 1.0351324081420898, | |
| "learning_rate": 7.94799600379813e-05, | |
| "loss": 5.3272, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 33.55434782608695, | |
| "grad_norm": 1.089340329170227, | |
| "learning_rate": 7.763841854293145e-05, | |
| "loss": 5.2996, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 34.19565217391305, | |
| "grad_norm": 1.3611856698989868, | |
| "learning_rate": 7.574121695112954e-05, | |
| "loss": 5.1266, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 34.84782608695652, | |
| "grad_norm": 1.2501380443572998, | |
| "learning_rate": 7.379217597409688e-05, | |
| "loss": 5.0434, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 35.48913043478261, | |
| "grad_norm": 1.057522177696228, | |
| "learning_rate": 7.179522072097774e-05, | |
| "loss": 4.9784, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 35.48913043478261, | |
| "eval_accuracy": 0.0011830630140928385, | |
| "eval_loss": 9.981914520263672, | |
| "eval_runtime": 106.635, | |
| "eval_samples_per_second": 2.832, | |
| "eval_steps_per_second": 0.178, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 36.130434782608695, | |
| "grad_norm": 1.1768474578857422, | |
| "learning_rate": 6.975437279389181e-05, | |
| "loss": 4.9012, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 36.78260869565217, | |
| "grad_norm": 1.1783802509307861, | |
| "learning_rate": 6.767374218896286e-05, | |
| "loss": 4.823, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 37.42391304347826, | |
| "grad_norm": 1.220082402229309, | |
| "learning_rate": 6.555751901933342e-05, | |
| "loss": 4.7149, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 38.06521739130435, | |
| "grad_norm": 1.3078495264053345, | |
| "learning_rate": 6.340996507683458e-05, | |
| "loss": 4.6413, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 38.71739130434783, | |
| "grad_norm": 1.2146966457366943, | |
| "learning_rate": 6.123540524930442e-05, | |
| "loss": 4.6684, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 38.71739130434783, | |
| "eval_accuracy": 0.0008952909295837696, | |
| "eval_loss": 10.014237403869629, | |
| "eval_runtime": 31.4866, | |
| "eval_samples_per_second": 9.591, | |
| "eval_steps_per_second": 0.603, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 39.391304347826086, | |
| "grad_norm": 1.090649962425232, | |
| "learning_rate": 5.903821881083942e-05, | |
| "loss": 4.7794, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 40.03260869565217, | |
| "grad_norm": 1.2051453590393066, | |
| "learning_rate": 5.682283060251932e-05, | |
| "loss": 4.4631, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 40.68478260869565, | |
| "grad_norm": 1.0512608289718628, | |
| "learning_rate": 5.4593702121365955e-05, | |
| "loss": 4.4119, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 41.32608695652174, | |
| "grad_norm": 0.9912136793136597, | |
| "learning_rate": 5.235532253548213e-05, | |
| "loss": 4.3377, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 41.97826086956522, | |
| "grad_norm": 1.0219991207122803, | |
| "learning_rate": 5.0112199643464376e-05, | |
| "loss": 4.3184, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 41.97826086956522, | |
| "eval_accuracy": 0.001019192799302952, | |
| "eval_loss": 10.048251152038574, | |
| "eval_runtime": 18.0166, | |
| "eval_samples_per_second": 16.762, | |
| "eval_steps_per_second": 1.055, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 42.619565217391305, | |
| "grad_norm": 0.9927310347557068, | |
| "learning_rate": 4.7868850796296495e-05, | |
| "loss": 4.331, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 43.26086956521739, | |
| "grad_norm": 0.9491915106773376, | |
| "learning_rate": 4.5629793800005945e-05, | |
| "loss": 4.1006, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 43.91304347826087, | |
| "grad_norm": 0.8775396347045898, | |
| "learning_rate": 4.339953781740363e-05, | |
| "loss": 4.177, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 44.55434782608695, | |
| "grad_norm": 0.9140155911445618, | |
| "learning_rate": 4.1182574287230224e-05, | |
| "loss": 4.0757, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 45.19565217391305, | |
| "grad_norm": 0.9226499199867249, | |
| "learning_rate": 3.898336787899612e-05, | |
| "loss": 4.1251, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 45.19565217391305, | |
| "eval_accuracy": 0.0008153542394423617, | |
| "eval_loss": 10.09643840789795, | |
| "eval_runtime": 21.8568, | |
| "eval_samples_per_second": 13.817, | |
| "eval_steps_per_second": 0.869, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 45.84782608695652, | |
| "grad_norm": 0.9727521538734436, | |
| "learning_rate": 3.680634750173137e-05, | |
| "loss": 4.0495, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 46.48913043478261, | |
| "grad_norm": 0.8741424083709717, | |
| "learning_rate": 3.4655897384752146e-05, | |
| "loss": 4.0195, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 47.130434782608695, | |
| "grad_norm": 0.9479995965957642, | |
| "learning_rate": 3.2536348248406534e-05, | |
| "loss": 3.9482, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 47.78260869565217, | |
| "grad_norm": 0.9046297073364258, | |
| "learning_rate": 3.0451968582579915e-05, | |
| "loss": 3.9166, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 48.42391304347826, | |
| "grad_norm": 0.861109733581543, | |
| "learning_rate": 2.840695605052458e-05, | |
| "loss": 3.909, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 48.42391304347826, | |
| "eval_accuracy": 0.0008992877640908401, | |
| "eval_loss": 10.132174491882324, | |
| "eval_runtime": 28.1979, | |
| "eval_samples_per_second": 10.71, | |
| "eval_steps_per_second": 0.674, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 49.06521739130435, | |
| "grad_norm": 0.8144867420196533, | |
| "learning_rate": 2.6405429035324403e-05, | |
| "loss": 3.9054, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 49.71739130434783, | |
| "grad_norm": 0.8467565178871155, | |
| "learning_rate": 2.4451418346019573e-05, | |
| "loss": 3.8543, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 50.358695652173914, | |
| "grad_norm": 0.8617934584617615, | |
| "learning_rate": 2.2548859100093407e-05, | |
| "loss": 3.8013, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "grad_norm": 1.1794334650039673, | |
| "learning_rate": 2.0701582798669676e-05, | |
| "loss": 3.803, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 51.65217391304348, | |
| "grad_norm": 0.861342191696167, | |
| "learning_rate": 1.8913309610379015e-05, | |
| "loss": 3.7535, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 51.65217391304348, | |
| "eval_accuracy": 0.0008753067570484176, | |
| "eval_loss": 10.158663749694824, | |
| "eval_runtime": 31.7351, | |
| "eval_samples_per_second": 9.516, | |
| "eval_steps_per_second": 0.599, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 52.29347826086956, | |
| "grad_norm": 0.8501729369163513, | |
| "learning_rate": 1.7187640879434553e-05, | |
| "loss": 3.7589, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 52.94565217391305, | |
| "grad_norm": 0.8114346861839294, | |
| "learning_rate": 1.552805187300389e-05, | |
| "loss": 3.8172, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 53.58695652173913, | |
| "grad_norm": 0.8227590918540955, | |
| "learning_rate": 1.3937884782483484e-05, | |
| "loss": 3.6933, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 54.22826086956522, | |
| "grad_norm": 0.7768607139587402, | |
| "learning_rate": 1.242034199277008e-05, | |
| "loss": 3.8079, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 54.880434782608695, | |
| "grad_norm": 0.8110019564628601, | |
| "learning_rate": 1.097847963308351e-05, | |
| "loss": 3.681, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 54.880434782608695, | |
| "eval_accuracy": 0.0008273447429635729, | |
| "eval_loss": 10.178533554077148, | |
| "eval_runtime": 31.6166, | |
| "eval_samples_per_second": 9.552, | |
| "eval_steps_per_second": 0.601, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 55.52173913043478, | |
| "grad_norm": 0.8320772647857666, | |
| "learning_rate": 9.615201422329406e-06, | |
| "loss": 3.6494, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 56.16304347826087, | |
| "grad_norm": 0.7713989019393921, | |
| "learning_rate": 8.333252821395526e-06, | |
| "loss": 3.7021, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 56.81521739130435, | |
| "grad_norm": 0.7743974924087524, | |
| "learning_rate": 7.135215504159115e-06, | |
| "loss": 3.7404, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 57.45652173913044, | |
| "grad_norm": 0.7438375353813171, | |
| "learning_rate": 6.023502158339078e-06, | |
| "loss": 3.6467, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 58.09782608695652, | |
| "grad_norm": 0.8585782051086426, | |
| "learning_rate": 5.000351626664207e-06, | |
| "loss": 3.688, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 58.09782608695652, | |
| "eval_accuracy": 0.0008273447429635729, | |
| "eval_loss": 10.187094688415527, | |
| "eval_runtime": 31.7031, | |
| "eval_samples_per_second": 9.526, | |
| "eval_steps_per_second": 0.599, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 58.75, | |
| "grad_norm": 0.7883967161178589, | |
| "learning_rate": 4.067824398141701e-06, | |
| "loss": 3.6471, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 59.391304347826086, | |
| "grad_norm": 0.7768418192863464, | |
| "learning_rate": 3.2277984585066366e-06, | |
| "loss": 3.6824, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 60.03260869565217, | |
| "grad_norm": 0.7814875245094299, | |
| "learning_rate": 2.4819655082085835e-06, | |
| "loss": 3.6767, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 60.68478260869565, | |
| "grad_norm": 0.8055542707443237, | |
| "learning_rate": 1.8318275555520237e-06, | |
| "loss": 3.5899, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 61.32608695652174, | |
| "grad_norm": 0.8155117630958557, | |
| "learning_rate": 1.2786938918515568e-06, | |
| "loss": 3.6685, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 61.32608695652174, | |
| "eval_accuracy": 0.0008233479084565024, | |
| "eval_loss": 10.19116497039795, | |
| "eval_runtime": 36.7079, | |
| "eval_samples_per_second": 8.227, | |
| "eval_steps_per_second": 0.518, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 61.97826086956522, | |
| "grad_norm": 0.8085272312164307, | |
| "learning_rate": 8.236784546933718e-07, | |
| "loss": 3.697, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 62.619565217391305, | |
| "grad_norm": 0.7857375741004944, | |
| "learning_rate": 4.676975846132692e-07, | |
| "loss": 3.6836, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 63.26086956521739, | |
| "grad_norm": 0.7917608618736267, | |
| "learning_rate": 2.1146817970871258e-07, | |
| "loss": 3.6163, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 63.91304347826087, | |
| "grad_norm": 0.7354781031608582, | |
| "learning_rate": 5.550625190150483e-08, | |
| "loss": 3.6652, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 64.55434782608695, | |
| "grad_norm": 0.7902089357376099, | |
| "learning_rate": 1.2588775841204658e-10, | |
| "loss": 3.6326, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 64.55434782608695, | |
| "eval_accuracy": 0.0008273447429635729, | |
| "eval_loss": 10.191176414489746, | |
| "eval_runtime": 37.3382, | |
| "eval_samples_per_second": 8.088, | |
| "eval_steps_per_second": 0.509, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 65, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2610102966336000.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |