| { |
| "best_metric": 0.6106137037277222, |
| "best_model_checkpoint": "tinybert_train_book_ent_15p_inv_cola/checkpoint-102", |
| "epoch": 8.0, |
| "eval_steps": 500, |
| "global_step": 272, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.8532902598381042, |
| "learning_rate": 4.9e-05, |
| "loss": 0.6137, |
| "step": 34 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6912751793861389, |
| "eval_loss": 0.6152212619781494, |
| "eval_matthews_correlation": 0.0, |
| "eval_runtime": 0.4041, |
| "eval_samples_per_second": 2580.839, |
| "eval_steps_per_second": 12.372, |
| "step": 34 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.9145865440368652, |
| "learning_rate": 4.8e-05, |
| "loss": 0.596, |
| "step": 68 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.690316379070282, |
| "eval_loss": 0.6138748526573181, |
| "eval_matthews_correlation": 0.013712849373137744, |
| "eval_runtime": 0.4213, |
| "eval_samples_per_second": 2475.667, |
| "eval_steps_per_second": 11.868, |
| "step": 68 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 1.5256273746490479, |
| "learning_rate": 4.7e-05, |
| "loss": 0.5732, |
| "step": 102 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.690316379070282, |
| "eval_loss": 0.6106137037277222, |
| "eval_matthews_correlation": 0.09689353339898876, |
| "eval_runtime": 0.4175, |
| "eval_samples_per_second": 2498.171, |
| "eval_steps_per_second": 11.976, |
| "step": 102 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 3.3182785511016846, |
| "learning_rate": 4.600000000000001e-05, |
| "loss": 0.5334, |
| "step": 136 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.6922339200973511, |
| "eval_loss": 0.6584149599075317, |
| "eval_matthews_correlation": 0.10411659709641652, |
| "eval_runtime": 0.3978, |
| "eval_samples_per_second": 2621.754, |
| "eval_steps_per_second": 12.568, |
| "step": 136 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 2.314908266067505, |
| "learning_rate": 4.5e-05, |
| "loss": 0.4908, |
| "step": 170 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.6548417806625366, |
| "eval_loss": 0.6714115738868713, |
| "eval_matthews_correlation": 0.10313429660779676, |
| "eval_runtime": 0.3957, |
| "eval_samples_per_second": 2635.572, |
| "eval_steps_per_second": 12.635, |
| "step": 170 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 3.4710166454315186, |
| "learning_rate": 4.4000000000000006e-05, |
| "loss": 0.443, |
| "step": 204 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.6711409687995911, |
| "eval_loss": 0.7121756672859192, |
| "eval_matthews_correlation": 0.16276809661834724, |
| "eval_runtime": 0.3963, |
| "eval_samples_per_second": 2632.172, |
| "eval_steps_per_second": 12.618, |
| "step": 204 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 2.8041465282440186, |
| "learning_rate": 4.3e-05, |
| "loss": 0.3959, |
| "step": 238 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.641418993473053, |
| "eval_loss": 0.7230662107467651, |
| "eval_matthews_correlation": 0.09815450223922367, |
| "eval_runtime": 0.3865, |
| "eval_samples_per_second": 2698.36, |
| "eval_steps_per_second": 12.936, |
| "step": 238 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 3.8785295486450195, |
| "learning_rate": 4.2e-05, |
| "loss": 0.3562, |
| "step": 272 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.6538830399513245, |
| "eval_loss": 0.8125724792480469, |
| "eval_matthews_correlation": 0.1294516933862166, |
| "eval_runtime": 0.4252, |
| "eval_samples_per_second": 2453.078, |
| "eval_steps_per_second": 11.76, |
| "step": 272 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 272, |
| "total_flos": 1793902531485696.0, |
| "train_loss": 0.5002556969137752, |
| "train_runtime": 50.5375, |
| "train_samples_per_second": 8460.05, |
| "train_steps_per_second": 33.638 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 1700, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 50, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.0 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 5 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1793902531485696.0, |
| "train_batch_size": 256, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
