| { | |
| "best_metric": 0.988673139158576, | |
| "best_model_checkpoint": "audio/train/checkpoint/hubert-large/superb_ks_42/checkpoint-9582", | |
| "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": 2.3518919944763184, | |
| "learning_rate": 3.1308703819661866e-05, | |
| "loss": 1.6086, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9826419535157399, | |
| "eval_loss": 0.09663528203964233, | |
| "eval_runtime": 25.3736, | |
| "eval_samples_per_second": 267.916, | |
| "eval_steps_per_second": 1.064, | |
| "step": 1597 | |
| }, | |
| { | |
| "epoch": 1.2523481527864746, | |
| "grad_norm": 1.2851642370224, | |
| "learning_rate": 4.859806581785292e-05, | |
| "loss": 0.3782, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.878522229179712, | |
| "grad_norm": 0.9166728854179382, | |
| "learning_rate": 4.51193209490016e-05, | |
| "loss": 0.2065, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9861724036481319, | |
| "eval_loss": 0.061149440705776215, | |
| "eval_runtime": 14.242, | |
| "eval_samples_per_second": 477.32, | |
| "eval_steps_per_second": 1.896, | |
| "step": 3194 | |
| }, | |
| { | |
| "epoch": 2.504696305572949, | |
| "grad_norm": 3.817800521850586, | |
| "learning_rate": 4.1640576080150286e-05, | |
| "loss": 0.1822, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9879376287143278, | |
| "eval_loss": 0.05412260442972183, | |
| "eval_runtime": 14.4843, | |
| "eval_samples_per_second": 469.334, | |
| "eval_steps_per_second": 1.864, | |
| "step": 4791 | |
| }, | |
| { | |
| "epoch": 3.1308703819661865, | |
| "grad_norm": 0.46756044030189514, | |
| "learning_rate": 3.8161831211298965e-05, | |
| "loss": 0.1565, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.757044458359424, | |
| "grad_norm": 2.2845380306243896, | |
| "learning_rate": 3.468308634244764e-05, | |
| "loss": 0.1529, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9869079140923801, | |
| "eval_loss": 0.05400605872273445, | |
| "eval_runtime": 14.6949, | |
| "eval_samples_per_second": 462.61, | |
| "eval_steps_per_second": 1.837, | |
| "step": 6388 | |
| }, | |
| { | |
| "epoch": 4.383218534752661, | |
| "grad_norm": 0.9110573530197144, | |
| "learning_rate": 3.120434147359633e-05, | |
| "loss": 0.1392, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9864666078258312, | |
| "eval_loss": 0.0533934086561203, | |
| "eval_runtime": 14.5906, | |
| "eval_samples_per_second": 465.917, | |
| "eval_steps_per_second": 1.851, | |
| "step": 7985 | |
| }, | |
| { | |
| "epoch": 5.009392611145898, | |
| "grad_norm": 2.609574317932129, | |
| "learning_rate": 2.7725596604745007e-05, | |
| "loss": 0.1331, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.635566687539136, | |
| "grad_norm": 0.1722152829170227, | |
| "learning_rate": 2.4246851735893692e-05, | |
| "loss": 0.1307, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.988673139158576, | |
| "eval_loss": 0.046409524977207184, | |
| "eval_runtime": 14.6338, | |
| "eval_samples_per_second": 464.542, | |
| "eval_steps_per_second": 1.845, | |
| "step": 9582 | |
| }, | |
| { | |
| "epoch": 6.261740763932373, | |
| "grad_norm": 0.5856819152832031, | |
| "learning_rate": 2.076810686704237e-05, | |
| "loss": 0.1221, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.88791484032561, | |
| "grad_norm": 1.7293404340744019, | |
| "learning_rate": 1.7289361998191053e-05, | |
| "loss": 0.1171, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9873492203589291, | |
| "eval_loss": 0.04979054629802704, | |
| "eval_runtime": 14.5666, | |
| "eval_samples_per_second": 466.685, | |
| "eval_steps_per_second": 1.854, | |
| "step": 11179 | |
| }, | |
| { | |
| "epoch": 7.514088916718848, | |
| "grad_norm": 0.8187317848205566, | |
| "learning_rate": 1.3810617129339735e-05, | |
| "loss": 0.1144, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9883789349808767, | |
| "eval_loss": 0.04705435782670975, | |
| "eval_runtime": 14.1821, | |
| "eval_samples_per_second": 479.337, | |
| "eval_steps_per_second": 1.904, | |
| "step": 12776 | |
| }, | |
| { | |
| "epoch": 8.140262993112085, | |
| "grad_norm": 0.5122739672660828, | |
| "learning_rate": 1.0331872260488417e-05, | |
| "loss": 0.114, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.766437069505322, | |
| "grad_norm": 0.05307776853442192, | |
| "learning_rate": 6.853127391637097e-06, | |
| "loss": 0.1104, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9879376287143278, | |
| "eval_loss": 0.045528609305620193, | |
| "eval_runtime": 14.768, | |
| "eval_samples_per_second": 460.319, | |
| "eval_steps_per_second": 1.828, | |
| "step": 14373 | |
| }, | |
| { | |
| "epoch": 9.39261114589856, | |
| "grad_norm": 2.4742355346679688, | |
| "learning_rate": 3.3743825227857787e-06, | |
| "loss": 0.1079, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9885260370697264, | |
| "eval_loss": 0.04639420658349991, | |
| "eval_runtime": 14.1401, | |
| "eval_samples_per_second": 480.761, | |
| "eval_steps_per_second": 1.909, | |
| "step": 15970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 15970, | |
| "total_flos": 1.548536697520896e+19, | |
| "train_loss": 0.2424131767258617, | |
| "train_runtime": 3138.2662, | |
| "train_samples_per_second": 162.81, | |
| "train_steps_per_second": 5.089 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15970, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.548536697520896e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |