| { | |
| "best_metric": 0.5263157894736842, | |
| "best_model_checkpoint": "videomae-base-sarco-aumentado/checkpoint-33", | |
| "epoch": 9.071875, | |
| "eval_steps": 500, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03125, | |
| "grad_norm": 7.657312870025635, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 1.414, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 10.452422142028809, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.2125, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09375, | |
| "grad_norm": 6.368149757385254, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.7603, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 1.5046850442886353, | |
| "eval_runtime": 9.1382, | |
| "eval_samples_per_second": 2.079, | |
| "eval_steps_per_second": 0.328, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.021875, | |
| "grad_norm": 17.073139190673828, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.4153, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.053125, | |
| "grad_norm": 8.040921211242676, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 0.1479, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.084375, | |
| "grad_norm": 0.2993019223213196, | |
| "learning_rate": 4.5138888888888894e-05, | |
| "loss": 0.0509, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 2.6908650398254395, | |
| "eval_runtime": 5.6622, | |
| "eval_samples_per_second": 3.356, | |
| "eval_steps_per_second": 0.53, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.0125, | |
| "grad_norm": 10.6167573928833, | |
| "learning_rate": 4.340277777777778e-05, | |
| "loss": 0.0109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.04375, | |
| "grad_norm": 1.5939055681228638, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.0043, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.075, | |
| "grad_norm": 0.037472404539585114, | |
| "learning_rate": 3.993055555555556e-05, | |
| "loss": 0.0422, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.123065948486328, | |
| "eval_runtime": 5.6279, | |
| "eval_samples_per_second": 3.376, | |
| "eval_steps_per_second": 0.533, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.003125, | |
| "grad_norm": 0.02211487852036953, | |
| "learning_rate": 3.8194444444444444e-05, | |
| "loss": 0.0025, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.034375, | |
| "grad_norm": 0.01639764942228794, | |
| "learning_rate": 3.6458333333333336e-05, | |
| "loss": 0.0016, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.065625, | |
| "grad_norm": 0.01938652992248535, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.0159, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.096875, | |
| "grad_norm": 0.017730316147208214, | |
| "learning_rate": 3.2986111111111115e-05, | |
| "loss": 0.0085, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.3051815032958984, | |
| "eval_runtime": 5.7488, | |
| "eval_samples_per_second": 3.305, | |
| "eval_steps_per_second": 0.522, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 4.025, | |
| "grad_norm": 0.017928821966052055, | |
| "learning_rate": 3.125e-05, | |
| "loss": 0.0096, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.05625, | |
| "grad_norm": 0.0493478998541832, | |
| "learning_rate": 2.951388888888889e-05, | |
| "loss": 0.0013, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0875, | |
| "grad_norm": 0.02256612479686737, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.0016, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.3200395107269287, | |
| "eval_runtime": 5.4706, | |
| "eval_samples_per_second": 3.473, | |
| "eval_steps_per_second": 0.548, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 5.015625, | |
| "grad_norm": 0.017535550519824028, | |
| "learning_rate": 2.604166666666667e-05, | |
| "loss": 0.0016, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.046875, | |
| "grad_norm": 0.012387892231345177, | |
| "learning_rate": 2.4305555555555558e-05, | |
| "loss": 0.001, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.078125, | |
| "grad_norm": 0.014036540873348713, | |
| "learning_rate": 2.2569444444444447e-05, | |
| "loss": 0.0011, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.3733017444610596, | |
| "eval_runtime": 5.6974, | |
| "eval_samples_per_second": 3.335, | |
| "eval_steps_per_second": 0.527, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 6.00625, | |
| "grad_norm": 0.011159941554069519, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.0009, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.0375, | |
| "grad_norm": 0.02007906697690487, | |
| "learning_rate": 1.9097222222222222e-05, | |
| "loss": 0.0009, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.06875, | |
| "grad_norm": 0.01185558270663023, | |
| "learning_rate": 1.736111111111111e-05, | |
| "loss": 0.0009, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "grad_norm": 0.012541444040834904, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 0.0008, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.3552279472351074, | |
| "eval_runtime": 5.687, | |
| "eval_samples_per_second": 3.341, | |
| "eval_steps_per_second": 0.528, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 7.028125, | |
| "grad_norm": 0.008836383931338787, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.0007, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.059375, | |
| "grad_norm": 0.010615321807563305, | |
| "learning_rate": 1.2152777777777779e-05, | |
| "loss": 0.0008, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.090625, | |
| "grad_norm": 0.010104900225996971, | |
| "learning_rate": 1.0416666666666668e-05, | |
| "loss": 0.0007, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 7.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.392703056335449, | |
| "eval_runtime": 5.807, | |
| "eval_samples_per_second": 3.272, | |
| "eval_steps_per_second": 0.517, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 8.01875, | |
| "grad_norm": 0.0091309929266572, | |
| "learning_rate": 8.680555555555556e-06, | |
| "loss": 0.0007, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 0.009017580188810825, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.0007, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.08125, | |
| "grad_norm": 0.009177467785775661, | |
| "learning_rate": 5.208333333333334e-06, | |
| "loss": 0.0007, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.103125, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.4122025966644287, | |
| "eval_runtime": 5.7143, | |
| "eval_samples_per_second": 3.325, | |
| "eval_steps_per_second": 0.525, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 9.009375, | |
| "grad_norm": 0.009887206368148327, | |
| "learning_rate": 3.4722222222222224e-06, | |
| "loss": 0.0007, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.040625, | |
| "grad_norm": 0.35840049386024475, | |
| "learning_rate": 1.7361111111111112e-06, | |
| "loss": 0.0009, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.071875, | |
| "grad_norm": 0.01002708449959755, | |
| "learning_rate": 0.0, | |
| "loss": 0.0007, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.071875, | |
| "eval_accuracy": 0.5263157894736842, | |
| "eval_loss": 3.4094326496124268, | |
| "eval_runtime": 6.4245, | |
| "eval_samples_per_second": 2.957, | |
| "eval_steps_per_second": 0.467, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.071875, | |
| "step": 320, | |
| "total_flos": 3.1451231201258373e+18, | |
| "train_loss": 0.12853713909717043, | |
| "train_runtime": 960.1181, | |
| "train_samples_per_second": 2.666, | |
| "train_steps_per_second": 0.333 | |
| }, | |
| { | |
| "epoch": 9.071875, | |
| "eval_accuracy": 0.3958333333333333, | |
| "eval_loss": 1.3436609506607056, | |
| "eval_runtime": 22.1033, | |
| "eval_samples_per_second": 2.172, | |
| "eval_steps_per_second": 0.271, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.071875, | |
| "eval_accuracy": 0.3958333333333333, | |
| "eval_loss": 1.3436609506607056, | |
| "eval_runtime": 14.4365, | |
| "eval_samples_per_second": 3.325, | |
| "eval_steps_per_second": 0.416, | |
| "step": 320 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 320, | |
| "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": 3.1451231201258373e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |