| { | |
| "best_metric": 0.9833774639599883, | |
| "best_model_checkpoint": "audio/train/checkpoint/hubert-base/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": 3.687915086746216, | |
| "learning_rate": 3.1308703819661866e-05, | |
| "loss": 1.5858, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9701382759635186, | |
| "eval_loss": 0.1746179759502411, | |
| "eval_runtime": 10.5438, | |
| "eval_samples_per_second": 644.74, | |
| "eval_steps_per_second": 2.561, | |
| "step": 1597 | |
| }, | |
| { | |
| "epoch": 1.2523481527864746, | |
| "grad_norm": 3.3637237548828125, | |
| "learning_rate": 4.859806581785292e-05, | |
| "loss": 0.4328, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.878522229179712, | |
| "grad_norm": 2.0145509243011475, | |
| "learning_rate": 4.51193209490016e-05, | |
| "loss": 0.2268, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9773462783171522, | |
| "eval_loss": 0.10376445204019547, | |
| "eval_runtime": 10.6724, | |
| "eval_samples_per_second": 636.969, | |
| "eval_steps_per_second": 2.53, | |
| "step": 3194 | |
| }, | |
| { | |
| "epoch": 2.504696305572949, | |
| "grad_norm": 9.941652297973633, | |
| "learning_rate": 4.1640576080150286e-05, | |
| "loss": 0.1937, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9796999117387467, | |
| "eval_loss": 0.08871696144342422, | |
| "eval_runtime": 6.7605, | |
| "eval_samples_per_second": 1005.553, | |
| "eval_steps_per_second": 3.994, | |
| "step": 4791 | |
| }, | |
| { | |
| "epoch": 3.1308703819661865, | |
| "grad_norm": 3.4196901321411133, | |
| "learning_rate": 3.8161831211298965e-05, | |
| "loss": 0.1745, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.757044458359424, | |
| "grad_norm": 10.562482833862305, | |
| "learning_rate": 3.468308634244764e-05, | |
| "loss": 0.163, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9826419535157399, | |
| "eval_loss": 0.07863254845142365, | |
| "eval_runtime": 6.5412, | |
| "eval_samples_per_second": 1039.264, | |
| "eval_steps_per_second": 4.128, | |
| "step": 6388 | |
| }, | |
| { | |
| "epoch": 4.383218534752661, | |
| "grad_norm": 8.417603492736816, | |
| "learning_rate": 3.120434147359633e-05, | |
| "loss": 0.1468, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9814651368049426, | |
| "eval_loss": 0.09819477796554565, | |
| "eval_runtime": 6.6232, | |
| "eval_samples_per_second": 1026.399, | |
| "eval_steps_per_second": 4.077, | |
| "step": 7985 | |
| }, | |
| { | |
| "epoch": 5.009392611145898, | |
| "grad_norm": 8.484418869018555, | |
| "learning_rate": 2.7725596604745007e-05, | |
| "loss": 0.1463, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.635566687539136, | |
| "grad_norm": 2.67555832862854, | |
| "learning_rate": 2.4246851735893692e-05, | |
| "loss": 0.1327, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9820535451603413, | |
| "eval_loss": 0.08876165747642517, | |
| "eval_runtime": 6.6465, | |
| "eval_samples_per_second": 1022.797, | |
| "eval_steps_per_second": 4.062, | |
| "step": 9582 | |
| }, | |
| { | |
| "epoch": 6.261740763932373, | |
| "grad_norm": 1.1208974123001099, | |
| "learning_rate": 2.076810686704237e-05, | |
| "loss": 0.1256, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.88791484032561, | |
| "grad_norm": 1.1888195276260376, | |
| "learning_rate": 1.7289361998191053e-05, | |
| "loss": 0.1175, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9811709326272433, | |
| "eval_loss": 0.09669487923383713, | |
| "eval_runtime": 6.767, | |
| "eval_samples_per_second": 1004.575, | |
| "eval_steps_per_second": 3.99, | |
| "step": 11179 | |
| }, | |
| { | |
| "epoch": 7.514088916718848, | |
| "grad_norm": 11.39383602142334, | |
| "learning_rate": 1.3810617129339735e-05, | |
| "loss": 0.1164, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9814651368049426, | |
| "eval_loss": 0.09223304688930511, | |
| "eval_runtime": 6.6066, | |
| "eval_samples_per_second": 1028.975, | |
| "eval_steps_per_second": 4.087, | |
| "step": 12776 | |
| }, | |
| { | |
| "epoch": 8.140262993112085, | |
| "grad_norm": 2.318204402923584, | |
| "learning_rate": 1.0331872260488417e-05, | |
| "loss": 0.1125, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.766437069505322, | |
| "grad_norm": 0.40608373284339905, | |
| "learning_rate": 6.853127391637097e-06, | |
| "loss": 0.1023, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9827890556045896, | |
| "eval_loss": 0.08754321187734604, | |
| "eval_runtime": 6.712, | |
| "eval_samples_per_second": 1012.816, | |
| "eval_steps_per_second": 4.023, | |
| "step": 14373 | |
| }, | |
| { | |
| "epoch": 9.39261114589856, | |
| "grad_norm": 0.3792332410812378, | |
| "learning_rate": 3.3743825227857787e-06, | |
| "loss": 0.0983, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9833774639599883, | |
| "eval_loss": 0.08568365126848221, | |
| "eval_runtime": 6.8598, | |
| "eval_samples_per_second": 990.99, | |
| "eval_steps_per_second": 3.936, | |
| "step": 15970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 15970, | |
| "total_flos": 4.6387626201984e+18, | |
| "train_loss": 0.2486723821612545, | |
| "train_runtime": 1368.9283, | |
| "train_samples_per_second": 373.241, | |
| "train_steps_per_second": 11.666 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15970, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 4.6387626201984e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |