| { | |
| "best_metric": 0.5532552003860474, | |
| "best_model_checkpoint": "bert_tiny_lda_5_v1_book_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": 0.9584577679634094, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.608, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.5934335589408875, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3362, | |
| "eval_samples_per_second": 3101.953, | |
| "eval_steps_per_second": 14.87, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.1856284141540527, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.5582, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7133269309997559, | |
| "eval_loss": 0.5757725238800049, | |
| "eval_matthews_correlation": 0.21056723021623536, | |
| "eval_runtime": 0.3487, | |
| "eval_samples_per_second": 2990.688, | |
| "eval_steps_per_second": 14.337, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.2397983074188232, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.4656, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7286673188209534, | |
| "eval_loss": 0.5532552003860474, | |
| "eval_matthews_correlation": 0.2911416522578436, | |
| "eval_runtime": 0.3619, | |
| "eval_samples_per_second": 2881.981, | |
| "eval_steps_per_second": 13.816, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.243818283081055, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.386, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7257909774780273, | |
| "eval_loss": 0.6198986768722534, | |
| "eval_matthews_correlation": 0.2783851428937992, | |
| "eval_runtime": 0.3305, | |
| "eval_samples_per_second": 3155.587, | |
| "eval_steps_per_second": 15.127, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 5.036900043487549, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3164, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7190795540809631, | |
| "eval_loss": 0.649246096611023, | |
| "eval_matthews_correlation": 0.27384707024880584, | |
| "eval_runtime": 0.3392, | |
| "eval_samples_per_second": 3075.234, | |
| "eval_steps_per_second": 14.742, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 4.896561145782471, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.2656, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7238734364509583, | |
| "eval_loss": 0.6995246410369873, | |
| "eval_matthews_correlation": 0.2903905003262487, | |
| "eval_runtime": 0.3342, | |
| "eval_samples_per_second": 3120.562, | |
| "eval_steps_per_second": 14.96, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 7.966293811798096, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2146, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7267497777938843, | |
| "eval_loss": 0.8111398816108704, | |
| "eval_matthews_correlation": 0.2862670482920985, | |
| "eval_runtime": 0.3314, | |
| "eval_samples_per_second": 3147.321, | |
| "eval_steps_per_second": 15.088, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.7810440063476562, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1798, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7238734364509583, | |
| "eval_loss": 0.7927041053771973, | |
| "eval_matthews_correlation": 0.2940303604806882, | |
| "eval_runtime": 0.3232, | |
| "eval_samples_per_second": 3227.352, | |
| "eval_steps_per_second": 15.471, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 1793902531485696.0, | |
| "train_loss": 0.374275638776667, | |
| "train_runtime": 49.5834, | |
| "train_samples_per_second": 8622.849, | |
| "train_steps_per_second": 34.286 | |
| } | |
| ], | |
| "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 | |
| } | |