| { | |
| "best_metric": 0.9833774639599883, | |
| "best_model_checkpoint": "audio/train/checkpoint/whisper-base/superb_ks_42/checkpoint-12776", | |
| "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": 47.73924255371094, | |
| "learning_rate": 3.1308703819661866e-05, | |
| "loss": 0.6909, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9651368049426302, | |
| "eval_loss": 0.157243549823761, | |
| "eval_runtime": 51.3965, | |
| "eval_samples_per_second": 132.266, | |
| "eval_steps_per_second": 0.525, | |
| "step": 1597 | |
| }, | |
| { | |
| "epoch": 1.2523481527864746, | |
| "grad_norm": 0.04125841706991196, | |
| "learning_rate": 4.859806581785292e-05, | |
| "loss": 0.1133, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.878522229179712, | |
| "grad_norm": 0.08289146423339844, | |
| "learning_rate": 4.51193209490016e-05, | |
| "loss": 0.0891, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9660194174757282, | |
| "eval_loss": 0.1596662998199463, | |
| "eval_runtime": 48.5884, | |
| "eval_samples_per_second": 139.91, | |
| "eval_steps_per_second": 0.556, | |
| "step": 3194 | |
| }, | |
| { | |
| "epoch": 2.504696305572949, | |
| "grad_norm": 0.05116074159741402, | |
| "learning_rate": 4.1640576080150286e-05, | |
| "loss": 0.0676, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9719035010297146, | |
| "eval_loss": 0.1304367631673813, | |
| "eval_runtime": 49.7204, | |
| "eval_samples_per_second": 136.724, | |
| "eval_steps_per_second": 0.543, | |
| "step": 4791 | |
| }, | |
| { | |
| "epoch": 3.1308703819661865, | |
| "grad_norm": 22.351160049438477, | |
| "learning_rate": 3.8161831211298965e-05, | |
| "loss": 0.0574, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.757044458359424, | |
| "grad_norm": 2.8176746368408203, | |
| "learning_rate": 3.468308634244764e-05, | |
| "loss": 0.0475, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.979552809649897, | |
| "eval_loss": 0.09990404546260834, | |
| "eval_runtime": 49.7998, | |
| "eval_samples_per_second": 136.506, | |
| "eval_steps_per_second": 0.542, | |
| "step": 6388 | |
| }, | |
| { | |
| "epoch": 4.383218534752661, | |
| "grad_norm": 0.7865376472473145, | |
| "learning_rate": 3.120434147359633e-05, | |
| "loss": 0.0433, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9798470138275963, | |
| "eval_loss": 0.10789525508880615, | |
| "eval_runtime": 47.3761, | |
| "eval_samples_per_second": 143.49, | |
| "eval_steps_per_second": 0.57, | |
| "step": 7985 | |
| }, | |
| { | |
| "epoch": 5.009392611145898, | |
| "grad_norm": 0.004570442717522383, | |
| "learning_rate": 2.7725596604745007e-05, | |
| "loss": 0.0392, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 5.635566687539136, | |
| "grad_norm": 0.0032151760533452034, | |
| "learning_rate": 2.4246851735893692e-05, | |
| "loss": 0.0284, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9802883200941453, | |
| "eval_loss": 0.10893493890762329, | |
| "eval_runtime": 46.9434, | |
| "eval_samples_per_second": 144.813, | |
| "eval_steps_per_second": 0.575, | |
| "step": 9582 | |
| }, | |
| { | |
| "epoch": 6.261740763932373, | |
| "grad_norm": 0.23338378965854645, | |
| "learning_rate": 2.076810686704237e-05, | |
| "loss": 0.0319, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 6.88791484032561, | |
| "grad_norm": 0.0019194560591131449, | |
| "learning_rate": 1.7289361998191053e-05, | |
| "loss": 0.0236, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9819064430714917, | |
| "eval_loss": 0.11618690937757492, | |
| "eval_runtime": 47.0803, | |
| "eval_samples_per_second": 144.391, | |
| "eval_steps_per_second": 0.573, | |
| "step": 11179 | |
| }, | |
| { | |
| "epoch": 7.514088916718848, | |
| "grad_norm": 0.0018724563997238874, | |
| "learning_rate": 1.3810617129339735e-05, | |
| "loss": 0.0193, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9833774639599883, | |
| "eval_loss": 0.1151697114109993, | |
| "eval_runtime": 47.0809, | |
| "eval_samples_per_second": 144.39, | |
| "eval_steps_per_second": 0.573, | |
| "step": 12776 | |
| }, | |
| { | |
| "epoch": 8.140262993112085, | |
| "grad_norm": 0.00741382222622633, | |
| "learning_rate": 1.0331872260488417e-05, | |
| "loss": 0.0162, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 8.766437069505322, | |
| "grad_norm": 0.004007210489362478, | |
| "learning_rate": 6.853127391637097e-06, | |
| "loss": 0.0111, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9820535451603413, | |
| "eval_loss": 0.12724293768405914, | |
| "eval_runtime": 46.9104, | |
| "eval_samples_per_second": 144.915, | |
| "eval_steps_per_second": 0.576, | |
| "step": 14373 | |
| }, | |
| { | |
| "epoch": 9.39261114589856, | |
| "grad_norm": 0.0008135234820656478, | |
| "learning_rate": 3.3743825227857787e-06, | |
| "loss": 0.0088, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9826419535157399, | |
| "eval_loss": 0.13061200082302094, | |
| "eval_runtime": 46.6644, | |
| "eval_samples_per_second": 145.678, | |
| "eval_steps_per_second": 0.579, | |
| "step": 15970 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 15970, | |
| "total_flos": 1.46834375382144e+19, | |
| "train_loss": 0.08098915571261736, | |
| "train_runtime": 8039.2826, | |
| "train_samples_per_second": 63.555, | |
| "train_steps_per_second": 1.986 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 15970, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.46834375382144e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |