| { | |
| "best_metric": 0.507537688442211, | |
| "best_model_checkpoint": "videomae-base-finetuned-ElderReact-Happiness11/checkpoint-156", | |
| "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": 3.6178653240203857, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.6756, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 4.96007776260376, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.6977, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.377594470977783, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.6977, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.4635678391959799, | |
| "eval_loss": 0.7047719359397888, | |
| "eval_runtime": 261.9257, | |
| "eval_samples_per_second": 3.039, | |
| "eval_steps_per_second": 0.191, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.699587345123291, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.7053, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 6.8290114402771, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.665, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 2.537061929702759, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.7055, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 1.5095937252044678, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.7198, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.47110552763819097, | |
| "eval_loss": 0.7133272290229797, | |
| "eval_runtime": 280.3033, | |
| "eval_samples_per_second": 2.84, | |
| "eval_steps_per_second": 0.178, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 2.1461029052734375, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6286, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.1844964027404785, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.6547, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 4.2716522216796875, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.6401, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 4.427234649658203, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.639, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.40703517587939697, | |
| "eval_loss": 0.8557814359664917, | |
| "eval_runtime": 261.857, | |
| "eval_samples_per_second": 3.04, | |
| "eval_steps_per_second": 0.191, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 2.5250039100646973, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.7011, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 4.220043659210205, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.602, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 7.634807109832764, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.6536, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 2.9651682376861572, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.664, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.507537688442211, | |
| "eval_loss": 0.6846373081207275, | |
| "eval_runtime": 252.5146, | |
| "eval_samples_per_second": 3.152, | |
| "eval_steps_per_second": 0.198, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 5.29097318649292, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.6102, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 4.258277416229248, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.6022, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 2.0270493030548096, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.6251, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 3.647946834564209, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6217, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.41457286432160806, | |
| "eval_loss": 0.8411366939544678, | |
| "eval_runtime": 254.7722, | |
| "eval_samples_per_second": 3.124, | |
| "eval_steps_per_second": 0.196, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 5.861187934875488, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.6029, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 4.177021026611328, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.5714, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 2.6523263454437256, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.6066, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 8.681379318237305, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.5826, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.42462311557788945, | |
| "eval_loss": 0.8426756262779236, | |
| "eval_runtime": 246.0667, | |
| "eval_samples_per_second": 3.235, | |
| "eval_steps_per_second": 0.203, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 5.220656871795654, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.5107, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 9.55013656616211, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.4856, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 4.11871862411499, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.5916, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 4.019587516784668, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.5881, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.435929648241206, | |
| "eval_loss": 0.8049066662788391, | |
| "eval_runtime": 263.5585, | |
| "eval_samples_per_second": 3.02, | |
| "eval_steps_per_second": 0.19, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 4.318753719329834, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.5706, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 4.975842475891113, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.5435, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 4.172918796539307, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.575, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 4.347151279449463, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.421, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.5050251256281407, | |
| "eval_loss": 0.778738796710968, | |
| "eval_runtime": 244.8077, | |
| "eval_samples_per_second": 3.252, | |
| "eval_steps_per_second": 0.204, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 6.215036392211914, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.4506, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 6.92239236831665, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.6033, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 2.6235239505767822, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.4765, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "grad_norm": 5.367151260375977, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.4674, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.47361809045226133, | |
| "eval_loss": 0.8280923962593079, | |
| "eval_runtime": 266.1328, | |
| "eval_samples_per_second": 2.991, | |
| "eval_steps_per_second": 0.188, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 3.808866500854492, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.5096, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 5.840443134307861, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.4805, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 6.295402526855469, | |
| "learning_rate": 0.0, | |
| "loss": 0.4775, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.4434673366834171, | |
| "eval_loss": 0.8615225553512573, | |
| "eval_runtime": 248.7642, | |
| "eval_samples_per_second": 3.2, | |
| "eval_steps_per_second": 0.201, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "step": 380, | |
| "total_flos": 7.475141472240992e+18, | |
| "train_loss": 0.5953672107897306, | |
| "train_runtime": 4834.4844, | |
| "train_samples_per_second": 1.258, | |
| "train_steps_per_second": 0.079 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.49866666666666665, | |
| "eval_loss": 0.7278424501419067, | |
| "eval_runtime": 243.6787, | |
| "eval_samples_per_second": 3.078, | |
| "eval_steps_per_second": 0.193, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.49866666666666665, | |
| "eval_loss": 0.7278424501419067, | |
| "eval_runtime": 236.8099, | |
| "eval_samples_per_second": 3.167, | |
| "eval_steps_per_second": 0.198, | |
| "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 | |
| } | |