| { | |
| "best_metric": 0.5767580270767212, | |
| "best_model_checkpoint": "bert_base_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": 3.1427481174468994, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6092, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6076905727386475, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.724, | |
| "eval_samples_per_second": 1440.604, | |
| "eval_steps_per_second": 6.906, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.963242769241333, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.604, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6922339200973511, | |
| "eval_loss": 0.5946319699287415, | |
| "eval_matthews_correlation": 0.0463559874942472, | |
| "eval_runtime": 0.6888, | |
| "eval_samples_per_second": 1514.119, | |
| "eval_steps_per_second": 7.258, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 3.547236442565918, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5599, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.715244472026825, | |
| "eval_loss": 0.5767580270767212, | |
| "eval_matthews_correlation": 0.23606681501621046, | |
| "eval_runtime": 0.7063, | |
| "eval_samples_per_second": 1476.798, | |
| "eval_steps_per_second": 7.08, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 4.889894485473633, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4932, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7219558954238892, | |
| "eval_loss": 0.5934705138206482, | |
| "eval_matthews_correlation": 0.2502797436163603, | |
| "eval_runtime": 0.6969, | |
| "eval_samples_per_second": 1496.573, | |
| "eval_steps_per_second": 7.174, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.201941967010498, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.3926, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7238734364509583, | |
| "eval_loss": 0.6084581017494202, | |
| "eval_matthews_correlation": 0.26948103360672326, | |
| "eval_runtime": 0.7302, | |
| "eval_samples_per_second": 1428.439, | |
| "eval_steps_per_second": 6.848, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 3.9843389987945557, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.3084, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7277085185050964, | |
| "eval_loss": 0.6839298009872437, | |
| "eval_matthews_correlation": 0.3134805838230471, | |
| "eval_runtime": 0.7112, | |
| "eval_samples_per_second": 1466.609, | |
| "eval_steps_per_second": 7.031, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.1595659255981445, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.2416, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.742090106010437, | |
| "eval_loss": 0.7191688418388367, | |
| "eval_matthews_correlation": 0.3343240846981291, | |
| "eval_runtime": 0.6869, | |
| "eval_samples_per_second": 1518.309, | |
| "eval_steps_per_second": 7.279, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.707186222076416, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.1883, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7382550239562988, | |
| "eval_loss": 0.8671514987945557, | |
| "eval_matthews_correlation": 0.31539041080291097, | |
| "eval_runtime": 0.7043, | |
| "eval_samples_per_second": 1480.959, | |
| "eval_steps_per_second": 7.1, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 8999450537533440.0, | |
| "train_loss": 0.4246412112432368, | |
| "train_runtime": 139.8899, | |
| "train_samples_per_second": 3056.331, | |
| "train_steps_per_second": 12.152 | |
| } | |
| ], | |
| "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": 8999450537533440.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |