| { | |
| "best_metric": 0.9839658723153869, | |
| "best_model_checkpoint": "audio/train/checkpoint/wav2vec2-large/superb_ks_42/checkpoint-15970", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 15970, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6261740763932373, | |
| "grad_norm": 7.050719738006592, | |
| "learning_rate": 3.1308703819661866e-05, | |
| "loss": 1.0646, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9624889673433362, | |
| "eval_loss": 0.1839100420475006, | |
| "eval_runtime": 14.3255, | |
| "eval_samples_per_second": 474.537, | |
| "eval_steps_per_second": 1.885, | |
| "step": 1597 | |
| }, | |
| { | |
| "epoch": 1.2523481527864746, | |
| "grad_norm": 9.074967384338379, | |
| "learning_rate": 4.859806581785292e-05, | |
| "loss": 0.4257, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.878522229179712, | |
| "grad_norm": 91.34820556640625, | |
| "learning_rate": 4.51193209490016e-05, | |
| "loss": 0.3751, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9646954986760812, | |
| "eval_loss": 0.19544720649719238, | |
| "eval_runtime": 13.9956, | |
| "eval_samples_per_second": 485.723, | |
| "eval_steps_per_second": 1.929, | |
| "step": 3194 | |
| }, | |
| { | |
| "epoch": 2.504696305572949, | |
| "grad_norm": 1.8326303958892822, | |
| "learning_rate": 4.1640576080150286e-05, | |
| "loss": 0.3156, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9744042365401588, | |
| "eval_loss": 0.13350164890289307, | |
| "eval_runtime": 14.2445, | |
| "eval_samples_per_second": 477.238, | |
| "eval_steps_per_second": 1.895, | |
| "step": 4791 | |
| }, | |
| { | |
| "epoch": 3.1308703819661865, | |
| "grad_norm": 2.554948329925537, | |
| "learning_rate": 3.8161831211298965e-05, | |
| "loss": 0.2839, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.757044458359424, | |
| "grad_norm": 1.011059045791626, | |
| "learning_rate": 3.468308634244764e-05, | |
| "loss": 0.257, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.979552809649897, | |
| "eval_loss": 0.10616285353899002, | |
| "eval_runtime": 14.1933, | |
| "eval_samples_per_second": 478.959, | |
| "eval_steps_per_second": 1.902, | |
| "step": 6388 | |
| }, | |
| { | |
| "epoch": 4.383218534752661, | |
| "grad_norm": 18.241968154907227, | |
| "learning_rate": 3.120434147359633e-05, | |
| "loss": 0.2386, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9801412180052956, | |
| "eval_loss": 0.10293199121952057, | |
| "eval_runtime": 21.533, | |
| "eval_samples_per_second": 315.701, | |
| "eval_steps_per_second": 1.254, | |
| "step": 7985 | |
| }, | |
| { | |
| "epoch": 5.009392611145898, | |
| "grad_norm": 22.520170211791992, | |
| "learning_rate": 2.7725596604745007e-05, | |
| "loss": 0.2205, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.635566687539136, | |
| "grad_norm": 2.018341302871704, | |
| "learning_rate": 2.4246851735893692e-05, | |
| "loss": 0.2085, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9814651368049426, | |
| "eval_loss": 0.1001678854227066, | |
| "eval_runtime": 14.0995, | |
| "eval_samples_per_second": 482.145, | |
| "eval_steps_per_second": 1.915, | |
| "step": 9582 | |
| }, | |
| { | |
| "epoch": 6.261740763932373, | |
| "grad_norm": 29.089778900146484, | |
| "learning_rate": 2.076810686704237e-05, | |
| "loss": 0.1905, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.88791484032561, | |
| "grad_norm": 2.23885178565979, | |
| "learning_rate": 1.7289361998191053e-05, | |
| "loss": 0.1715, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.981759340982642, | |
| "eval_loss": 0.10314840078353882, | |
| "eval_runtime": 14.0909, | |
| "eval_samples_per_second": 482.44, | |
| "eval_steps_per_second": 1.916, | |
| "step": 11179 | |
| }, | |
| { | |
| "epoch": 7.514088916718848, | |
| "grad_norm": 0.31174391508102417, | |
| "learning_rate": 1.3810617129339735e-05, | |
| "loss": 0.1575, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9819064430714917, | |
| "eval_loss": 0.09382686764001846, | |
| "eval_runtime": 14.1853, | |
| "eval_samples_per_second": 479.227, | |
| "eval_steps_per_second": 1.903, | |
| "step": 12776 | |
| }, | |
| { | |
| "epoch": 8.140262993112085, | |
| "grad_norm": 2.9018185138702393, | |
| "learning_rate": 1.0331872260488417e-05, | |
| "loss": 0.147, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.766437069505322, | |
| "grad_norm": 3.0589239597320557, | |
| "learning_rate": 6.853127391637097e-06, | |
| "loss": 0.1332, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9830832597822889, | |
| "eval_loss": 0.08958739787340164, | |
| "eval_runtime": 14.0026, | |
| "eval_samples_per_second": 485.481, | |
| "eval_steps_per_second": 1.928, | |
| "step": 14373 | |
| }, | |
| { | |
| "epoch": 9.39261114589856, | |
| "grad_norm": 4.1695756912231445, | |
| "learning_rate": 3.3743825227857787e-06, | |
| "loss": 0.1288, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9839658723153869, | |
| "eval_loss": 0.08564967662096024, | |
| "eval_runtime": 13.9588, | |
| "eval_samples_per_second": 487.004, | |
| "eval_steps_per_second": 1.934, | |
| "step": 15970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 15970, | |
| "total_flos": 1.548488981447424e+19, | |
| "train_loss": 0.27784167333327014, | |
| "train_runtime": 3003.8643, | |
| "train_samples_per_second": 170.094, | |
| "train_steps_per_second": 5.316 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15970, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.548488981447424e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |