| { | |
| "best_metric": 0.35570469798657717, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-783", | |
| "epoch": 30.026442307692307, | |
| "eval_steps": 500, | |
| "global_step": 832, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01201923076923077, | |
| "grad_norm": 9.181777954101562, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 4.2532, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02403846153846154, | |
| "grad_norm": 6.192462921142578, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 4.2583, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03245192307692308, | |
| "eval_accuracy": 0.015659955257270694, | |
| "eval_loss": 4.225129127502441, | |
| "eval_runtime": 113.6438, | |
| "eval_samples_per_second": 3.933, | |
| "eval_steps_per_second": 0.123, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.0036057692307692, | |
| "grad_norm": 6.222796440124512, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 4.2505, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.015625, | |
| "grad_norm": 3.965486526489258, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 4.2489, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0276442307692308, | |
| "grad_norm": 5.0857744216918945, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 4.2374, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0324519230769231, | |
| "eval_accuracy": 0.013422818791946308, | |
| "eval_loss": 4.226736068725586, | |
| "eval_runtime": 113.905, | |
| "eval_samples_per_second": 3.924, | |
| "eval_steps_per_second": 0.123, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.0072115384615383, | |
| "grad_norm": 3.5929408073425293, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 4.2403, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.019230769230769, | |
| "grad_norm": 3.6417300701141357, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 4.255, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 3.2748305797576904, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 4.2678, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.032451923076923, | |
| "eval_accuracy": 0.013422818791946308, | |
| "eval_loss": 4.2263360023498535, | |
| "eval_runtime": 115.125, | |
| "eval_samples_per_second": 3.883, | |
| "eval_steps_per_second": 0.122, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.0108173076923075, | |
| "grad_norm": 3.1406173706054688, | |
| "learning_rate": 4.959893048128342e-05, | |
| "loss": 4.2151, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0228365384615383, | |
| "grad_norm": 2.8441436290740967, | |
| "learning_rate": 4.8930481283422465e-05, | |
| "loss": 4.2537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.032451923076923, | |
| "eval_accuracy": 0.015659955257270694, | |
| "eval_loss": 4.221240997314453, | |
| "eval_runtime": 115.863, | |
| "eval_samples_per_second": 3.858, | |
| "eval_steps_per_second": 0.121, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.002403846153846, | |
| "grad_norm": 2.7364041805267334, | |
| "learning_rate": 4.8262032085561496e-05, | |
| "loss": 4.2568, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.014423076923077, | |
| "grad_norm": 2.71181058883667, | |
| "learning_rate": 4.759358288770054e-05, | |
| "loss": 4.2288, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.0264423076923075, | |
| "grad_norm": 2.7206692695617676, | |
| "learning_rate": 4.6925133689839576e-05, | |
| "loss": 4.2401, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.032451923076923, | |
| "eval_accuracy": 0.015659955257270694, | |
| "eval_loss": 4.206515312194824, | |
| "eval_runtime": 114.2862, | |
| "eval_samples_per_second": 3.911, | |
| "eval_steps_per_second": 0.122, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.006009615384615, | |
| "grad_norm": 2.6117472648620605, | |
| "learning_rate": 4.625668449197861e-05, | |
| "loss": 4.2192, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.018028846153846, | |
| "grad_norm": 2.90580677986145, | |
| "learning_rate": 4.558823529411765e-05, | |
| "loss": 4.2204, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.030048076923077, | |
| "grad_norm": 2.9660141468048096, | |
| "learning_rate": 4.491978609625669e-05, | |
| "loss": 4.2519, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.032451923076923, | |
| "eval_accuracy": 0.017897091722595078, | |
| "eval_loss": 4.207516193389893, | |
| "eval_runtime": 117.7024, | |
| "eval_samples_per_second": 3.798, | |
| "eval_steps_per_second": 0.119, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 6.009615384615385, | |
| "grad_norm": 2.8973159790039062, | |
| "learning_rate": 4.4251336898395724e-05, | |
| "loss": 4.2112, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.021634615384615, | |
| "grad_norm": 2.734694480895996, | |
| "learning_rate": 4.358288770053476e-05, | |
| "loss": 4.2198, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.032451923076923, | |
| "eval_accuracy": 0.013422818791946308, | |
| "eval_loss": 4.205463409423828, | |
| "eval_runtime": 116.7575, | |
| "eval_samples_per_second": 3.828, | |
| "eval_steps_per_second": 0.12, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 7.001201923076923, | |
| "grad_norm": 2.421506881713867, | |
| "learning_rate": 4.29144385026738e-05, | |
| "loss": 4.2292, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 7.013221153846154, | |
| "grad_norm": 2.82832670211792, | |
| "learning_rate": 4.224598930481284e-05, | |
| "loss": 4.2019, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.025240384615385, | |
| "grad_norm": 3.0022928714752197, | |
| "learning_rate": 4.157754010695187e-05, | |
| "loss": 4.2111, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.032451923076923, | |
| "eval_accuracy": 0.017897091722595078, | |
| "eval_loss": 4.19577693939209, | |
| "eval_runtime": 119.1446, | |
| "eval_samples_per_second": 3.752, | |
| "eval_steps_per_second": 0.118, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 8.004807692307692, | |
| "grad_norm": 2.6406443119049072, | |
| "learning_rate": 4.0909090909090915e-05, | |
| "loss": 4.2068, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.016826923076923, | |
| "grad_norm": 3.169625759124756, | |
| "learning_rate": 4.024064171122995e-05, | |
| "loss": 4.1759, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.028846153846153, | |
| "grad_norm": 2.8210370540618896, | |
| "learning_rate": 3.957219251336899e-05, | |
| "loss": 4.1871, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.032451923076923, | |
| "eval_accuracy": 0.04697986577181208, | |
| "eval_loss": 4.147375583648682, | |
| "eval_runtime": 117.849, | |
| "eval_samples_per_second": 3.793, | |
| "eval_steps_per_second": 0.119, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 9.008413461538462, | |
| "grad_norm": 3.140075206756592, | |
| "learning_rate": 3.8903743315508025e-05, | |
| "loss": 4.1314, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 9.020432692307692, | |
| "grad_norm": 3.0806028842926025, | |
| "learning_rate": 3.8235294117647055e-05, | |
| "loss": 4.0778, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.032451923076923, | |
| "grad_norm": 12.136305809020996, | |
| "learning_rate": 3.75668449197861e-05, | |
| "loss": 4.0891, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.032451923076923, | |
| "eval_accuracy": 0.0447427293064877, | |
| "eval_loss": 4.032660007476807, | |
| "eval_runtime": 116.5729, | |
| "eval_samples_per_second": 3.835, | |
| "eval_steps_per_second": 0.12, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 10.01201923076923, | |
| "grad_norm": 4.812855243682861, | |
| "learning_rate": 3.6898395721925136e-05, | |
| "loss": 3.9545, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 10.024038461538462, | |
| "grad_norm": 5.218036651611328, | |
| "learning_rate": 3.622994652406417e-05, | |
| "loss": 3.7963, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 10.032451923076923, | |
| "eval_accuracy": 0.08277404921700224, | |
| "eval_loss": 3.82175874710083, | |
| "eval_runtime": 115.9952, | |
| "eval_samples_per_second": 3.854, | |
| "eval_steps_per_second": 0.121, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 11.00360576923077, | |
| "grad_norm": 4.765311241149902, | |
| "learning_rate": 3.556149732620321e-05, | |
| "loss": 3.7306, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 11.015625, | |
| "grad_norm": 6.956701755523682, | |
| "learning_rate": 3.489304812834225e-05, | |
| "loss": 3.5563, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 11.02764423076923, | |
| "grad_norm": 5.394344329833984, | |
| "learning_rate": 3.4224598930481284e-05, | |
| "loss": 3.4787, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 11.032451923076923, | |
| "eval_accuracy": 0.11185682326621924, | |
| "eval_loss": 3.7061688899993896, | |
| "eval_runtime": 113.6071, | |
| "eval_samples_per_second": 3.935, | |
| "eval_steps_per_second": 0.123, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 12.007211538461538, | |
| "grad_norm": 5.70036506652832, | |
| "learning_rate": 3.355614973262032e-05, | |
| "loss": 3.4821, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 12.01923076923077, | |
| "grad_norm": 6.370093822479248, | |
| "learning_rate": 3.288770053475936e-05, | |
| "loss": 3.3756, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 12.03125, | |
| "grad_norm": 5.8564133644104, | |
| "learning_rate": 3.22192513368984e-05, | |
| "loss": 3.1883, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 12.032451923076923, | |
| "eval_accuracy": 0.11185682326621924, | |
| "eval_loss": 3.58866810798645, | |
| "eval_runtime": 112.5771, | |
| "eval_samples_per_second": 3.971, | |
| "eval_steps_per_second": 0.124, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 13.010817307692308, | |
| "grad_norm": 6.884145736694336, | |
| "learning_rate": 3.155080213903743e-05, | |
| "loss": 3.1542, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 13.022836538461538, | |
| "grad_norm": 7.7231059074401855, | |
| "learning_rate": 3.0882352941176475e-05, | |
| "loss": 3.0045, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 13.032451923076923, | |
| "eval_accuracy": 0.14317673378076062, | |
| "eval_loss": 3.380981206893921, | |
| "eval_runtime": 113.3509, | |
| "eval_samples_per_second": 3.944, | |
| "eval_steps_per_second": 0.124, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 14.002403846153847, | |
| "grad_norm": 6.160039901733398, | |
| "learning_rate": 3.0213903743315508e-05, | |
| "loss": 2.9704, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 14.014423076923077, | |
| "grad_norm": 8.257513046264648, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 2.7723, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 14.026442307692308, | |
| "grad_norm": 6.512996196746826, | |
| "learning_rate": 2.8877005347593582e-05, | |
| "loss": 2.8045, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 14.032451923076923, | |
| "eval_accuracy": 0.21923937360178972, | |
| "eval_loss": 3.2211973667144775, | |
| "eval_runtime": 116.514, | |
| "eval_samples_per_second": 3.836, | |
| "eval_steps_per_second": 0.12, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 15.006009615384615, | |
| "grad_norm": 6.627754211425781, | |
| "learning_rate": 2.8208556149732622e-05, | |
| "loss": 2.7247, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 15.018028846153847, | |
| "grad_norm": 6.555057048797607, | |
| "learning_rate": 2.754010695187166e-05, | |
| "loss": 2.5789, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 15.030048076923077, | |
| "grad_norm": 8.675689697265625, | |
| "learning_rate": 2.68716577540107e-05, | |
| "loss": 2.5344, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 15.032451923076923, | |
| "eval_accuracy": 0.15883668903803133, | |
| "eval_loss": 3.270237922668457, | |
| "eval_runtime": 114.1519, | |
| "eval_samples_per_second": 3.916, | |
| "eval_steps_per_second": 0.123, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 16.009615384615383, | |
| "grad_norm": 7.248364448547363, | |
| "learning_rate": 2.6203208556149733e-05, | |
| "loss": 2.3791, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 16.021634615384617, | |
| "grad_norm": 9.409223556518555, | |
| "learning_rate": 2.5534759358288773e-05, | |
| "loss": 2.3725, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 16.032451923076923, | |
| "eval_accuracy": 0.14093959731543623, | |
| "eval_loss": 3.35996413230896, | |
| "eval_runtime": 113.8728, | |
| "eval_samples_per_second": 3.925, | |
| "eval_steps_per_second": 0.123, | |
| "step": 459 | |
| }, | |
| { | |
| "epoch": 17.001201923076923, | |
| "grad_norm": 7.834836006164551, | |
| "learning_rate": 2.4866310160427807e-05, | |
| "loss": 2.2722, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 17.013221153846153, | |
| "grad_norm": 7.643532752990723, | |
| "learning_rate": 2.4197860962566847e-05, | |
| "loss": 2.1434, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 17.025240384615383, | |
| "grad_norm": 8.425023078918457, | |
| "learning_rate": 2.3529411764705884e-05, | |
| "loss": 2.2074, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 17.032451923076923, | |
| "eval_accuracy": 0.2371364653243848, | |
| "eval_loss": 2.9730582237243652, | |
| "eval_runtime": 113.5853, | |
| "eval_samples_per_second": 3.935, | |
| "eval_steps_per_second": 0.123, | |
| "step": 486 | |
| }, | |
| { | |
| "epoch": 18.004807692307693, | |
| "grad_norm": 7.840888500213623, | |
| "learning_rate": 2.286096256684492e-05, | |
| "loss": 2.0823, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 18.016826923076923, | |
| "grad_norm": 9.385211944580078, | |
| "learning_rate": 2.2192513368983957e-05, | |
| "loss": 1.9511, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 18.028846153846153, | |
| "grad_norm": 8.72028636932373, | |
| "learning_rate": 2.1524064171122994e-05, | |
| "loss": 2.1094, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 18.032451923076923, | |
| "eval_accuracy": 0.26174496644295303, | |
| "eval_loss": 2.8679935932159424, | |
| "eval_runtime": 113.6275, | |
| "eval_samples_per_second": 3.934, | |
| "eval_steps_per_second": 0.123, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 19.00841346153846, | |
| "grad_norm": 9.284242630004883, | |
| "learning_rate": 2.0855614973262035e-05, | |
| "loss": 2.0278, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 19.020432692307693, | |
| "grad_norm": 6.913205623626709, | |
| "learning_rate": 2.018716577540107e-05, | |
| "loss": 1.8727, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 19.032451923076923, | |
| "grad_norm": 21.923686981201172, | |
| "learning_rate": 1.951871657754011e-05, | |
| "loss": 1.9839, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 19.032451923076923, | |
| "eval_accuracy": 0.27069351230425054, | |
| "eval_loss": 2.8359620571136475, | |
| "eval_runtime": 115.0445, | |
| "eval_samples_per_second": 3.885, | |
| "eval_steps_per_second": 0.122, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 20.01201923076923, | |
| "grad_norm": 8.244462013244629, | |
| "learning_rate": 1.8850267379679145e-05, | |
| "loss": 1.7584, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 20.02403846153846, | |
| "grad_norm": 8.496162414550781, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.7354, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 20.032451923076923, | |
| "eval_accuracy": 0.28187919463087246, | |
| "eval_loss": 2.7890186309814453, | |
| "eval_runtime": 111.5479, | |
| "eval_samples_per_second": 4.007, | |
| "eval_steps_per_second": 0.126, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 21.00360576923077, | |
| "grad_norm": 7.818769931793213, | |
| "learning_rate": 1.7513368983957222e-05, | |
| "loss": 1.701, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 21.015625, | |
| "grad_norm": 9.921280860900879, | |
| "learning_rate": 1.684491978609626e-05, | |
| "loss": 1.648, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 21.02764423076923, | |
| "grad_norm": 8.053401947021484, | |
| "learning_rate": 1.6176470588235296e-05, | |
| "loss": 1.6843, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 21.032451923076923, | |
| "eval_accuracy": 0.29977628635346754, | |
| "eval_loss": 2.728635787963867, | |
| "eval_runtime": 114.938, | |
| "eval_samples_per_second": 3.889, | |
| "eval_steps_per_second": 0.122, | |
| "step": 594 | |
| }, | |
| { | |
| "epoch": 22.00721153846154, | |
| "grad_norm": 10.640948295593262, | |
| "learning_rate": 1.5508021390374333e-05, | |
| "loss": 1.4562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 22.01923076923077, | |
| "grad_norm": 9.333779335021973, | |
| "learning_rate": 1.4839572192513372e-05, | |
| "loss": 1.5091, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 22.03125, | |
| "grad_norm": 9.433292388916016, | |
| "learning_rate": 1.4171122994652408e-05, | |
| "loss": 1.6266, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 22.032451923076923, | |
| "eval_accuracy": 0.2841163310961969, | |
| "eval_loss": 2.806154489517212, | |
| "eval_runtime": 115.4195, | |
| "eval_samples_per_second": 3.873, | |
| "eval_steps_per_second": 0.121, | |
| "step": 621 | |
| }, | |
| { | |
| "epoch": 23.010817307692307, | |
| "grad_norm": 8.609794616699219, | |
| "learning_rate": 1.3502673796791445e-05, | |
| "loss": 1.6214, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 23.02283653846154, | |
| "grad_norm": 6.943145275115967, | |
| "learning_rate": 1.2834224598930484e-05, | |
| "loss": 1.4083, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 23.032451923076923, | |
| "eval_accuracy": 0.2595078299776286, | |
| "eval_loss": 2.8204569816589355, | |
| "eval_runtime": 115.0186, | |
| "eval_samples_per_second": 3.886, | |
| "eval_steps_per_second": 0.122, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 24.002403846153847, | |
| "grad_norm": 10.106169700622559, | |
| "learning_rate": 1.2165775401069519e-05, | |
| "loss": 1.3264, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 24.014423076923077, | |
| "grad_norm": 10.27568244934082, | |
| "learning_rate": 1.1497326203208558e-05, | |
| "loss": 1.3662, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 24.026442307692307, | |
| "grad_norm": 10.36124038696289, | |
| "learning_rate": 1.0828877005347594e-05, | |
| "loss": 1.4422, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 24.032451923076923, | |
| "eval_accuracy": 0.30648769574944074, | |
| "eval_loss": 2.6406848430633545, | |
| "eval_runtime": 116.1998, | |
| "eval_samples_per_second": 3.847, | |
| "eval_steps_per_second": 0.12, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 25.006009615384617, | |
| "grad_norm": 8.807711601257324, | |
| "learning_rate": 1.0160427807486631e-05, | |
| "loss": 1.3912, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 25.018028846153847, | |
| "grad_norm": 8.629064559936523, | |
| "learning_rate": 9.49197860962567e-06, | |
| "loss": 1.2904, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 25.030048076923077, | |
| "grad_norm": 9.190649032592773, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 1.3897, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 25.032451923076923, | |
| "eval_accuracy": 0.34675615212527966, | |
| "eval_loss": 2.5948002338409424, | |
| "eval_runtime": 118.4696, | |
| "eval_samples_per_second": 3.773, | |
| "eval_steps_per_second": 0.118, | |
| "step": 702 | |
| }, | |
| { | |
| "epoch": 26.009615384615383, | |
| "grad_norm": 10.941615104675293, | |
| "learning_rate": 8.155080213903744e-06, | |
| "loss": 1.2647, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 26.021634615384617, | |
| "grad_norm": 9.282670021057129, | |
| "learning_rate": 7.4866310160427806e-06, | |
| "loss": 1.3906, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 26.032451923076923, | |
| "eval_accuracy": 0.31543624161073824, | |
| "eval_loss": 2.629518985748291, | |
| "eval_runtime": 120.316, | |
| "eval_samples_per_second": 3.715, | |
| "eval_steps_per_second": 0.116, | |
| "step": 729 | |
| }, | |
| { | |
| "epoch": 27.001201923076923, | |
| "grad_norm": 9.254827499389648, | |
| "learning_rate": 6.818181818181818e-06, | |
| "loss": 1.3257, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 27.013221153846153, | |
| "grad_norm": 7.69126558303833, | |
| "learning_rate": 6.149732620320856e-06, | |
| "loss": 1.2581, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 27.025240384615383, | |
| "grad_norm": 10.26952075958252, | |
| "learning_rate": 5.481283422459893e-06, | |
| "loss": 1.2291, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 27.032451923076923, | |
| "eval_accuracy": 0.3378076062639821, | |
| "eval_loss": 2.5539445877075195, | |
| "eval_runtime": 116.6007, | |
| "eval_samples_per_second": 3.834, | |
| "eval_steps_per_second": 0.12, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 28.004807692307693, | |
| "grad_norm": 11.07676887512207, | |
| "learning_rate": 4.812834224598931e-06, | |
| "loss": 1.29, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 28.016826923076923, | |
| "grad_norm": 8.940054893493652, | |
| "learning_rate": 4.144385026737968e-06, | |
| "loss": 1.1383, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 28.028846153846153, | |
| "grad_norm": 9.17213249206543, | |
| "learning_rate": 3.4759358288770056e-06, | |
| "loss": 1.3166, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 28.032451923076923, | |
| "eval_accuracy": 0.35570469798657717, | |
| "eval_loss": 2.519993305206299, | |
| "eval_runtime": 116.4901, | |
| "eval_samples_per_second": 3.837, | |
| "eval_steps_per_second": 0.12, | |
| "step": 783 | |
| }, | |
| { | |
| "epoch": 29.00841346153846, | |
| "grad_norm": 11.884153366088867, | |
| "learning_rate": 2.807486631016043e-06, | |
| "loss": 1.232, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 29.020432692307693, | |
| "grad_norm": 9.171028137207031, | |
| "learning_rate": 2.1390374331550802e-06, | |
| "loss": 1.1733, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 29.032451923076923, | |
| "grad_norm": 24.7429256439209, | |
| "learning_rate": 1.4705882352941177e-06, | |
| "loss": 1.2619, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 29.032451923076923, | |
| "eval_accuracy": 0.35570469798657717, | |
| "eval_loss": 2.530792236328125, | |
| "eval_runtime": 118.4418, | |
| "eval_samples_per_second": 3.774, | |
| "eval_steps_per_second": 0.118, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 30.01201923076923, | |
| "grad_norm": 8.288737297058105, | |
| "learning_rate": 8.021390374331552e-07, | |
| "loss": 1.1824, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 30.02403846153846, | |
| "grad_norm": 8.749896049499512, | |
| "learning_rate": 1.3368983957219251e-07, | |
| "loss": 1.1393, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 30.026442307692307, | |
| "eval_accuracy": 0.3243847874720358, | |
| "eval_loss": 2.5855681896209717, | |
| "eval_runtime": 118.0615, | |
| "eval_samples_per_second": 3.786, | |
| "eval_steps_per_second": 0.119, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 30.026442307692307, | |
| "step": 832, | |
| "total_flos": 3.2147153612960956e+19, | |
| "train_loss": 2.754230235058528, | |
| "train_runtime": 10713.0029, | |
| "train_samples_per_second": 2.485, | |
| "train_steps_per_second": 0.078 | |
| }, | |
| { | |
| "epoch": 30.026442307692307, | |
| "eval_accuracy": 0.35570469798657717, | |
| "eval_loss": 2.5194525718688965, | |
| "eval_runtime": 114.2775, | |
| "eval_samples_per_second": 3.912, | |
| "eval_steps_per_second": 0.123, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 30.026442307692307, | |
| "eval_accuracy": 0.35570469798657717, | |
| "eval_loss": 2.520230293273926, | |
| "eval_runtime": 112.9248, | |
| "eval_samples_per_second": 3.958, | |
| "eval_steps_per_second": 0.124, | |
| "step": 832 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 832, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2147153612960956e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |