| { | |
| "best_metric": 0.8285714285714286, | |
| "best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-450", | |
| "epoch": 3.25, | |
| "eval_steps": 500, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 14.608237266540527, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 2.2453, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 36.968116760253906, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 2.3136, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 17.145219802856445, | |
| "learning_rate": 2.5e-05, | |
| "loss": 2.42, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 20.452388763427734, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 2.3047, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 15.896721839904785, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 2.2122, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 14.993762969970703, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2581, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 16.3026065826416, | |
| "learning_rate": 4.9074074074074075e-05, | |
| "loss": 2.0639, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 13.365699768066406, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 2.1623, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 15.710684776306152, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 2.0486, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 20.03302001953125, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.9578, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 17.960296630859375, | |
| "learning_rate": 4.5370370370370374e-05, | |
| "loss": 1.952, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 15.012153625488281, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 1.623, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 18.308734893798828, | |
| "learning_rate": 4.351851851851852e-05, | |
| "loss": 1.7624, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 14.50885009765625, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 1.2651, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 24.735700607299805, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.3279, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_accuracy": 0.42857142857142855, | |
| "eval_loss": 1.2386213541030884, | |
| "eval_runtime": 29.1642, | |
| "eval_samples_per_second": 2.4, | |
| "eval_steps_per_second": 1.2, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 17.132925033569336, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 1.3581, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 21.23714828491211, | |
| "learning_rate": 3.981481481481482e-05, | |
| "loss": 1.1455, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 18.7020206451416, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 1.1061, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 14.506175994873047, | |
| "learning_rate": 3.7962962962962964e-05, | |
| "loss": 1.1518, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 29.114465713500977, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.9828, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 10.247823715209961, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 1.003, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 8.87232780456543, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.5946, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 160.70228576660156, | |
| "learning_rate": 3.425925925925926e-05, | |
| "loss": 1.1998, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 12.714564323425293, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.9956, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 3.044830560684204, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.5824, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.5050694346427917, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.5335, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 16.457557678222656, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.7113, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 16.745141983032227, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.397, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 4.463345050811768, | |
| "learning_rate": 2.8703703703703706e-05, | |
| "loss": 0.7122, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 12.840511322021484, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4469, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_accuracy": 0.7857142857142857, | |
| "eval_loss": 0.46059101819992065, | |
| "eval_runtime": 29.3265, | |
| "eval_samples_per_second": 2.387, | |
| "eval_steps_per_second": 1.193, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 25.240575790405273, | |
| "learning_rate": 2.6851851851851855e-05, | |
| "loss": 0.6132, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 3.6886823177337646, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.3257, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 7.413472652435303, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3551, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.6936644911766052, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.2338, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.7576755285263062, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.2495, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 2.146125078201294, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.4222, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 48.200714111328125, | |
| "learning_rate": 2.1296296296296296e-05, | |
| "loss": 0.6491, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.2533012628555298, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.3478, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 3.1355950832366943, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.3471, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 32.68257522583008, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.3896, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "grad_norm": 40.63040542602539, | |
| "learning_rate": 1.7592592592592595e-05, | |
| "loss": 0.423, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 11.917498588562012, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7307, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 0.4221107065677643, | |
| "learning_rate": 1.574074074074074e-05, | |
| "loss": 0.327, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "grad_norm": 0.340420126914978, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.4035, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 0.12714390456676483, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2142, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "eval_accuracy": 0.8285714285714286, | |
| "eval_loss": 0.38029590249061584, | |
| "eval_runtime": 29.2401, | |
| "eval_samples_per_second": 2.394, | |
| "eval_steps_per_second": 1.197, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 0.8629240393638611, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1078, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 0.08954589068889618, | |
| "learning_rate": 1.2037037037037037e-05, | |
| "loss": 0.238, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 0.12774558365345, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.0945, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "grad_norm": 0.49261391162872314, | |
| "learning_rate": 1.0185185185185185e-05, | |
| "loss": 0.2428, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 0.0893869623541832, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.2159, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 3.983721971511841, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.39, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.7854453325271606, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.2665, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 2.8955111503601074, | |
| "learning_rate": 6.481481481481481e-06, | |
| "loss": 0.0859, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "grad_norm": 15.494441032409668, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1552, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 0.1666683405637741, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0635, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 0.389921635389328, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.3509, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.4067876636981964, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.1328, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 0.7461154460906982, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.3664, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "grad_norm": 8.207443237304688, | |
| "learning_rate": 9.259259259259259e-07, | |
| "loss": 0.0469, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 0.5032423138618469, | |
| "learning_rate": 0.0, | |
| "loss": 0.0459, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.8285714285714286, | |
| "eval_loss": 0.3240524232387543, | |
| "eval_runtime": 29.9304, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 1.169, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "step": 600, | |
| "total_flos": 1.495384188125184e+18, | |
| "train_loss": 0.8612027465800445, | |
| "train_runtime": 1212.3004, | |
| "train_samples_per_second": 0.99, | |
| "train_steps_per_second": 0.495 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.832258064516129, | |
| "eval_loss": 0.48749518394470215, | |
| "eval_runtime": 67.379, | |
| "eval_samples_per_second": 2.3, | |
| "eval_steps_per_second": 1.158, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "eval_accuracy": 0.832258064516129, | |
| "eval_loss": 0.48749518394470215, | |
| "eval_runtime": 67.3551, | |
| "eval_samples_per_second": 2.301, | |
| "eval_steps_per_second": 1.158, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 1.495384188125184e+18, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |