| { | |
| "best_metric": 0.6106081008911133, | |
| "best_model_checkpoint": "bert_tiny_lda_100_v1_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": 0.9724855422973633, | |
| "learning_rate": 4.9e-05, | |
| "loss": 0.6154, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6165938973426819, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3572, | |
| "eval_samples_per_second": 2919.986, | |
| "eval_steps_per_second": 13.998, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.8256316781044006, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6048, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6912751793861389, | |
| "eval_loss": 0.6132587790489197, | |
| "eval_matthews_correlation": 0.0, | |
| "eval_runtime": 0.3489, | |
| "eval_samples_per_second": 2989.719, | |
| "eval_steps_per_second": 14.332, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.2026365995407104, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.5957, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6941514611244202, | |
| "eval_loss": 0.6106081008911133, | |
| "eval_matthews_correlation": 0.07380739515541786, | |
| "eval_runtime": 0.3351, | |
| "eval_samples_per_second": 3112.325, | |
| "eval_steps_per_second": 14.92, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4783580303192139, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5772, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.690316379070282, | |
| "eval_loss": 0.6260800361633301, | |
| "eval_matthews_correlation": 0.027408946028538616, | |
| "eval_runtime": 0.3689, | |
| "eval_samples_per_second": 2827.512, | |
| "eval_steps_per_second": 13.555, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.2147879600524902, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.5353, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6692233681678772, | |
| "eval_loss": 0.6314420104026794, | |
| "eval_matthews_correlation": 0.1251591159068797, | |
| "eval_runtime": 0.3358, | |
| "eval_samples_per_second": 3105.641, | |
| "eval_steps_per_second": 14.888, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.8101205825805664, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 0.496, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6260786056518555, | |
| "eval_loss": 0.6528063416481018, | |
| "eval_matthews_correlation": 0.09159056426356416, | |
| "eval_runtime": 0.3519, | |
| "eval_samples_per_second": 2963.966, | |
| "eval_steps_per_second": 14.209, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.8958512544631958, | |
| "learning_rate": 4.3e-05, | |
| "loss": 0.4518, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6251198649406433, | |
| "eval_loss": 0.6883458495140076, | |
| "eval_matthews_correlation": 0.08365265177329322, | |
| "eval_runtime": 0.3357, | |
| "eval_samples_per_second": 3107.316, | |
| "eval_steps_per_second": 14.896, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 3.256547212600708, | |
| "learning_rate": 4.2e-05, | |
| "loss": 0.4203, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.640460193157196, | |
| "eval_loss": 0.7734943628311157, | |
| "eval_matthews_correlation": 0.06813518347888069, | |
| "eval_runtime": 0.3283, | |
| "eval_samples_per_second": 3177.046, | |
| "eval_steps_per_second": 15.23, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 272, | |
| "total_flos": 1793902531485696.0, | |
| "train_loss": 0.5370645768502179, | |
| "train_runtime": 46.6261, | |
| "train_samples_per_second": 9169.749, | |
| "train_steps_per_second": 36.46 | |
| } | |
| ], | |
| "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 | |
| } | |