| { | |
| "best_metric": 0.9309045226130653, | |
| "best_model_checkpoint": "videomae-base-finetuned-ElderReact-Fear11/checkpoint-39", | |
| "epoch": 9.076315789473684, | |
| "eval_steps": 500, | |
| "global_step": 380, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 3.863062858581543, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.5556, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.5027570724487305, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.5024, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.4324733018875122, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.4132, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.25437450408935547, | |
| "eval_runtime": 266.7732, | |
| "eval_samples_per_second": 2.984, | |
| "eval_steps_per_second": 0.187, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.0784482955932617, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.3849, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 4.857975959777832, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.4374, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 2.003520965576172, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.4102, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 3.905989646911621, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.4058, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.2649548649787903, | |
| "eval_runtime": 312.2869, | |
| "eval_samples_per_second": 2.549, | |
| "eval_steps_per_second": 0.16, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 3.0998950004577637, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.397, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.9156787395477295, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.4688, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 0.8537417054176331, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.3165, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.6496905088424683, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.351, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.30776447057724, | |
| "eval_runtime": 291.3205, | |
| "eval_samples_per_second": 2.732, | |
| "eval_steps_per_second": 0.172, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 2.174421787261963, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.4136, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 1.2753701210021973, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.3649, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 2.671156644821167, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.4788, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 2.175607204437256, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.399, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.25298213958740234, | |
| "eval_runtime": 271.3874, | |
| "eval_samples_per_second": 2.933, | |
| "eval_steps_per_second": 0.184, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 2.933234930038452, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.3368, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 4.285318851470947, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.3704, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 2.213531732559204, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.4282, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 1.1714284420013428, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3695, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.24894379079341888, | |
| "eval_runtime": 286.1426, | |
| "eval_samples_per_second": 2.782, | |
| "eval_steps_per_second": 0.175, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 3.546438217163086, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.3161, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 2.6896326541900635, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.4168, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 1.6038317680358887, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.2991, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 4.019348621368408, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.3783, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.23861204087734222, | |
| "eval_runtime": 287.4515, | |
| "eval_samples_per_second": 2.769, | |
| "eval_steps_per_second": 0.174, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 2.982759952545166, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.308, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 1.170196294784546, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.2539, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 1.875527262687683, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.3984, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 2.0295588970184326, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.3474, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.928391959798995, | |
| "eval_loss": 0.2565610408782959, | |
| "eval_runtime": 281.7088, | |
| "eval_samples_per_second": 2.826, | |
| "eval_steps_per_second": 0.177, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 1.923046350479126, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.3105, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 1.6540353298187256, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.2194, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 2.46889591217041, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.3648, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 3.0250182151794434, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.3682, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.9309045226130653, | |
| "eval_loss": 0.23805934190750122, | |
| "eval_runtime": 289.0701, | |
| "eval_samples_per_second": 2.754, | |
| "eval_steps_per_second": 0.173, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 2.378732442855835, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.3726, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 3.59675931930542, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.2489, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 1.277233362197876, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.2707, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "grad_norm": 4.423985481262207, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.2805, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.9296482412060302, | |
| "eval_loss": 0.2403843104839325, | |
| "eval_runtime": 280.2511, | |
| "eval_samples_per_second": 2.84, | |
| "eval_steps_per_second": 0.178, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 3.3136565685272217, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.2588, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 2.6917359828948975, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.2183, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 3.4215519428253174, | |
| "learning_rate": 0.0, | |
| "loss": 0.3692, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.928391959798995, | |
| "eval_loss": 0.2460562288761139, | |
| "eval_runtime": 305.8945, | |
| "eval_samples_per_second": 2.602, | |
| "eval_steps_per_second": 0.163, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "step": 380, | |
| "total_flos": 7.475141472240992e+18, | |
| "train_loss": 0.36326786844353925, | |
| "train_runtime": 5259.7489, | |
| "train_samples_per_second": 1.156, | |
| "train_steps_per_second": 0.072 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.908, | |
| "eval_loss": 0.30357953906059265, | |
| "eval_runtime": 230.1824, | |
| "eval_samples_per_second": 3.258, | |
| "eval_steps_per_second": 0.204, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.908, | |
| "eval_loss": 0.30357953906059265, | |
| "eval_runtime": 210.0413, | |
| "eval_samples_per_second": 3.571, | |
| "eval_steps_per_second": 0.224, | |
| "step": 380 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 380, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 7.475141472240992e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |