|
{ |
|
"best_metric": 0.8561643835616438, |
|
"best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-dcsass-shoplifting-subset/checkpoint-780", |
|
"epoch": 4.194871794871795, |
|
"eval_steps": 500, |
|
"global_step": 780, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.01282051282051282, |
|
"grad_norm": 6.756944179534912, |
|
"learning_rate": 6.41025641025641e-06, |
|
"loss": 0.7103, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.02564102564102564, |
|
"grad_norm": 7.49935245513916, |
|
"learning_rate": 1.282051282051282e-05, |
|
"loss": 0.5933, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.038461538461538464, |
|
"grad_norm": 8.1102294921875, |
|
"learning_rate": 1.923076923076923e-05, |
|
"loss": 0.391, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.05128205128205128, |
|
"grad_norm": 14.223252296447754, |
|
"learning_rate": 2.564102564102564e-05, |
|
"loss": 0.5162, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.0641025641025641, |
|
"grad_norm": 10.161632537841797, |
|
"learning_rate": 3.205128205128206e-05, |
|
"loss": 0.5433, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.07692307692307693, |
|
"grad_norm": 7.198577404022217, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.7253, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.08974358974358974, |
|
"grad_norm": 15.455668449401855, |
|
"learning_rate": 4.4871794871794874e-05, |
|
"loss": 0.6042, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.10256410256410256, |
|
"grad_norm": 5.9158935546875, |
|
"learning_rate": 4.985754985754986e-05, |
|
"loss": 0.4834, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.11538461538461539, |
|
"grad_norm": 13.751437187194824, |
|
"learning_rate": 4.9145299145299147e-05, |
|
"loss": 0.6672, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.1282051282051282, |
|
"grad_norm": 6.230504989624023, |
|
"learning_rate": 4.8433048433048433e-05, |
|
"loss": 0.561, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.14102564102564102, |
|
"grad_norm": 0.061720337718725204, |
|
"learning_rate": 4.772079772079772e-05, |
|
"loss": 0.4631, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.15384615384615385, |
|
"grad_norm": 14.445475578308105, |
|
"learning_rate": 4.700854700854701e-05, |
|
"loss": 0.5993, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.16666666666666666, |
|
"grad_norm": 5.024906158447266, |
|
"learning_rate": 4.62962962962963e-05, |
|
"loss": 0.6052, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.1794871794871795, |
|
"grad_norm": 5.089183807373047, |
|
"learning_rate": 4.558404558404559e-05, |
|
"loss": 0.5345, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.19230769230769232, |
|
"grad_norm": 0.20361727476119995, |
|
"learning_rate": 4.4871794871794874e-05, |
|
"loss": 0.3314, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.2012820512820513, |
|
"eval_accuracy": 0.821917808219178, |
|
"eval_loss": 0.6146280169487, |
|
"eval_runtime": 86.3385, |
|
"eval_samples_per_second": 1.691, |
|
"eval_steps_per_second": 0.429, |
|
"step": 157 |
|
}, |
|
{ |
|
"epoch": 1.0038461538461538, |
|
"grad_norm": 0.9299275875091553, |
|
"learning_rate": 4.415954415954416e-05, |
|
"loss": 0.5923, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 1.0166666666666666, |
|
"grad_norm": 1.9701108932495117, |
|
"learning_rate": 4.344729344729345e-05, |
|
"loss": 0.6272, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 1.0294871794871794, |
|
"grad_norm": 7.168660640716553, |
|
"learning_rate": 4.2735042735042735e-05, |
|
"loss": 0.3787, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 1.0423076923076924, |
|
"grad_norm": 17.313304901123047, |
|
"learning_rate": 4.202279202279202e-05, |
|
"loss": 0.4095, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 1.0551282051282052, |
|
"grad_norm": 3.225602865219116, |
|
"learning_rate": 4.131054131054131e-05, |
|
"loss": 0.4833, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.067948717948718, |
|
"grad_norm": 11.712457656860352, |
|
"learning_rate": 4.05982905982906e-05, |
|
"loss": 0.6347, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 1.0807692307692307, |
|
"grad_norm": 15.161521911621094, |
|
"learning_rate": 3.988603988603989e-05, |
|
"loss": 0.4419, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 1.0935897435897437, |
|
"grad_norm": 11.030887603759766, |
|
"learning_rate": 3.9173789173789176e-05, |
|
"loss": 0.5758, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 1.1064102564102565, |
|
"grad_norm": 5.54761266708374, |
|
"learning_rate": 3.846153846153846e-05, |
|
"loss": 0.6464, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 1.1192307692307693, |
|
"grad_norm": 8.304089546203613, |
|
"learning_rate": 3.774928774928775e-05, |
|
"loss": 0.3947, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.132051282051282, |
|
"grad_norm": 0.35803475975990295, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.4621, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 1.1448717948717948, |
|
"grad_norm": 19.831405639648438, |
|
"learning_rate": 3.6324786324786323e-05, |
|
"loss": 0.8061, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 1.1576923076923076, |
|
"grad_norm": 0.5348687767982483, |
|
"learning_rate": 3.561253561253561e-05, |
|
"loss": 0.4076, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.1705128205128206, |
|
"grad_norm": 6.216036796569824, |
|
"learning_rate": 3.4900284900284904e-05, |
|
"loss": 0.403, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 1.1833333333333333, |
|
"grad_norm": 1.6393040418624878, |
|
"learning_rate": 3.418803418803419e-05, |
|
"loss": 0.4308, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.1961538461538461, |
|
"grad_norm": 5.89309549331665, |
|
"learning_rate": 3.347578347578348e-05, |
|
"loss": 0.3263, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 1.2012820512820512, |
|
"eval_accuracy": 0.8287671232876712, |
|
"eval_loss": 0.6184389591217041, |
|
"eval_runtime": 44.0336, |
|
"eval_samples_per_second": 3.316, |
|
"eval_steps_per_second": 0.84, |
|
"step": 314 |
|
}, |
|
{ |
|
"epoch": 2.0076923076923077, |
|
"grad_norm": 8.897279739379883, |
|
"learning_rate": 3.2763532763532764e-05, |
|
"loss": 0.736, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 2.0205128205128204, |
|
"grad_norm": 11.958976745605469, |
|
"learning_rate": 3.205128205128206e-05, |
|
"loss": 0.0545, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 2.033333333333333, |
|
"grad_norm": 0.10042566061019897, |
|
"learning_rate": 3.133903133903134e-05, |
|
"loss": 0.3544, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 2.046153846153846, |
|
"grad_norm": 3.468107223510742, |
|
"learning_rate": 3.0626780626780625e-05, |
|
"loss": 0.4753, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 2.0589743589743588, |
|
"grad_norm": 0.6979026794433594, |
|
"learning_rate": 2.9914529914529915e-05, |
|
"loss": 0.1313, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 2.071794871794872, |
|
"grad_norm": 3.0566658973693848, |
|
"learning_rate": 2.9202279202279202e-05, |
|
"loss": 0.5758, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 2.0846153846153848, |
|
"grad_norm": 0.26063987612724304, |
|
"learning_rate": 2.8490028490028492e-05, |
|
"loss": 0.2311, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 2.0974358974358975, |
|
"grad_norm": 3.1704676151275635, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 0.3121, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 2.1102564102564103, |
|
"grad_norm": 0.06660736352205276, |
|
"learning_rate": 2.706552706552707e-05, |
|
"loss": 0.2649, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.123076923076923, |
|
"grad_norm": 0.8849107027053833, |
|
"learning_rate": 2.6353276353276356e-05, |
|
"loss": 0.4632, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 2.135897435897436, |
|
"grad_norm": 16.48140525817871, |
|
"learning_rate": 2.564102564102564e-05, |
|
"loss": 0.5191, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 2.1487179487179486, |
|
"grad_norm": 2.628391742706299, |
|
"learning_rate": 2.492877492877493e-05, |
|
"loss": 0.405, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 2.1615384615384614, |
|
"grad_norm": 1.1029433012008667, |
|
"learning_rate": 2.4216524216524217e-05, |
|
"loss": 0.3149, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 2.174358974358974, |
|
"grad_norm": 8.070186614990234, |
|
"learning_rate": 2.3504273504273504e-05, |
|
"loss": 0.5851, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 2.1871794871794874, |
|
"grad_norm": 12.600333213806152, |
|
"learning_rate": 2.2792022792022794e-05, |
|
"loss": 0.3032, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 2.2, |
|
"grad_norm": 1.988337755203247, |
|
"learning_rate": 2.207977207977208e-05, |
|
"loss": 0.6332, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 2.2012820512820515, |
|
"eval_accuracy": 0.8424657534246576, |
|
"eval_loss": 0.4275156855583191, |
|
"eval_runtime": 44.8316, |
|
"eval_samples_per_second": 3.257, |
|
"eval_steps_per_second": 0.825, |
|
"step": 471 |
|
}, |
|
{ |
|
"epoch": 3.0115384615384615, |
|
"grad_norm": 0.7842563986778259, |
|
"learning_rate": 2.1367521367521368e-05, |
|
"loss": 0.2056, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 3.0243589743589743, |
|
"grad_norm": 9.073537826538086, |
|
"learning_rate": 2.0655270655270654e-05, |
|
"loss": 0.2923, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 3.037179487179487, |
|
"grad_norm": 9.187246322631836, |
|
"learning_rate": 1.9943019943019945e-05, |
|
"loss": 0.2992, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 3.05, |
|
"grad_norm": 20.043611526489258, |
|
"learning_rate": 1.923076923076923e-05, |
|
"loss": 0.5296, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 3.0628205128205126, |
|
"grad_norm": 28.867929458618164, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.1522, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 3.075641025641026, |
|
"grad_norm": 13.749585151672363, |
|
"learning_rate": 1.7806267806267805e-05, |
|
"loss": 0.5946, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 3.0884615384615386, |
|
"grad_norm": 3.9784131050109863, |
|
"learning_rate": 1.7094017094017095e-05, |
|
"loss": 0.7594, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 3.1012820512820514, |
|
"grad_norm": 0.4377283453941345, |
|
"learning_rate": 1.6381766381766382e-05, |
|
"loss": 0.0527, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 3.114102564102564, |
|
"grad_norm": 0.44730284810066223, |
|
"learning_rate": 1.566951566951567e-05, |
|
"loss": 0.3361, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 3.126923076923077, |
|
"grad_norm": 2.1688125133514404, |
|
"learning_rate": 1.4957264957264958e-05, |
|
"loss": 0.4073, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 3.1397435897435897, |
|
"grad_norm": 6.609060287475586, |
|
"learning_rate": 1.4245014245014246e-05, |
|
"loss": 0.5241, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 3.1525641025641025, |
|
"grad_norm": 18.088504791259766, |
|
"learning_rate": 1.3532763532763535e-05, |
|
"loss": 0.1817, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 3.1653846153846152, |
|
"grad_norm": 3.1018402576446533, |
|
"learning_rate": 1.282051282051282e-05, |
|
"loss": 0.1476, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 3.178205128205128, |
|
"grad_norm": 0.2694937586784363, |
|
"learning_rate": 1.2108262108262108e-05, |
|
"loss": 0.1166, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 3.1910256410256412, |
|
"grad_norm": 0.2130882292985916, |
|
"learning_rate": 1.1396011396011397e-05, |
|
"loss": 0.0405, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 3.2012820512820515, |
|
"eval_accuracy": 0.8424657534246576, |
|
"eval_loss": 0.76870197057724, |
|
"eval_runtime": 44.2253, |
|
"eval_samples_per_second": 3.301, |
|
"eval_steps_per_second": 0.837, |
|
"step": 628 |
|
}, |
|
{ |
|
"epoch": 4.0025641025641026, |
|
"grad_norm": 37.76756286621094, |
|
"learning_rate": 1.0683760683760684e-05, |
|
"loss": 0.4644, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 4.015384615384615, |
|
"grad_norm": 0.223415806889534, |
|
"learning_rate": 9.971509971509972e-06, |
|
"loss": 0.5616, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 4.028205128205128, |
|
"grad_norm": 0.11729208379983902, |
|
"learning_rate": 9.259259259259259e-06, |
|
"loss": 0.0667, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 4.041025641025641, |
|
"grad_norm": 3.586672782897949, |
|
"learning_rate": 8.547008547008548e-06, |
|
"loss": 0.1252, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 4.053846153846154, |
|
"grad_norm": 20.754961013793945, |
|
"learning_rate": 7.834757834757835e-06, |
|
"loss": 0.2335, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 4.066666666666666, |
|
"grad_norm": 1.776369571685791, |
|
"learning_rate": 7.122507122507123e-06, |
|
"loss": 0.1772, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 4.079487179487179, |
|
"grad_norm": 0.9859512448310852, |
|
"learning_rate": 6.41025641025641e-06, |
|
"loss": 0.1654, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 4.092307692307692, |
|
"grad_norm": 9.265326499938965, |
|
"learning_rate": 5.6980056980056985e-06, |
|
"loss": 0.1738, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 4.105128205128205, |
|
"grad_norm": 17.748327255249023, |
|
"learning_rate": 4.985754985754986e-06, |
|
"loss": 0.3303, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 4.1179487179487175, |
|
"grad_norm": 0.5790041089057922, |
|
"learning_rate": 4.273504273504274e-06, |
|
"loss": 0.2181, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 4.130769230769231, |
|
"grad_norm": 0.2577972710132599, |
|
"learning_rate": 3.5612535612535615e-06, |
|
"loss": 0.0101, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 4.143589743589744, |
|
"grad_norm": 27.089906692504883, |
|
"learning_rate": 2.8490028490028492e-06, |
|
"loss": 0.1569, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 4.156410256410257, |
|
"grad_norm": 0.044243063777685165, |
|
"learning_rate": 2.136752136752137e-06, |
|
"loss": 0.1688, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 4.1692307692307695, |
|
"grad_norm": 13.373856544494629, |
|
"learning_rate": 1.4245014245014246e-06, |
|
"loss": 0.0592, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 4.182051282051282, |
|
"grad_norm": 3.7931740283966064, |
|
"learning_rate": 7.122507122507123e-07, |
|
"loss": 0.2266, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 4.194871794871795, |
|
"grad_norm": 6.501523971557617, |
|
"learning_rate": 0.0, |
|
"loss": 0.1579, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 4.194871794871795, |
|
"eval_accuracy": 0.8561643835616438, |
|
"eval_loss": 0.5852294564247131, |
|
"eval_runtime": 44.3219, |
|
"eval_samples_per_second": 3.294, |
|
"eval_steps_per_second": 0.835, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 4.194871794871795, |
|
"step": 780, |
|
"total_flos": 3.8777530024360673e+18, |
|
"train_loss": 0.3928673116060404, |
|
"train_runtime": 2196.6027, |
|
"train_samples_per_second": 1.42, |
|
"train_steps_per_second": 0.355 |
|
}, |
|
{ |
|
"epoch": 4.194871794871795, |
|
"eval_accuracy": 0.8874172185430463, |
|
"eval_loss": 0.3803229033946991, |
|
"eval_runtime": 88.7221, |
|
"eval_samples_per_second": 1.702, |
|
"eval_steps_per_second": 0.428, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 4.194871794871795, |
|
"eval_accuracy": 0.8874172185430463, |
|
"eval_loss": 0.38032299280166626, |
|
"eval_runtime": 51.2971, |
|
"eval_samples_per_second": 2.944, |
|
"eval_steps_per_second": 0.741, |
|
"step": 780 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 780, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 9223372036854775807, |
|
"save_steps": 500, |
|
"total_flos": 3.8777530024360673e+18, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|