| { | |
| "best_metric": 0.47051816449999595, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-4276", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 4276, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.23386342376052385, | |
| "grad_norm": 4.184101104736328, | |
| "learning_rate": 1.7853399770993998e-05, | |
| "loss": 0.593, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4677268475210477, | |
| "grad_norm": 2.0551953315734863, | |
| "learning_rate": 1.548933730131789e-05, | |
| "loss": 0.5753, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7015902712815716, | |
| "grad_norm": 16.489328384399414, | |
| "learning_rate": 1.3125274831641774e-05, | |
| "loss": 0.5518, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.9354536950420954, | |
| "grad_norm": 47.566402435302734, | |
| "learning_rate": 1.0761212361965661e-05, | |
| "loss": 0.5419, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.6548950672149658, | |
| "eval_matthews_correlation": 0.4413232484146929, | |
| "eval_runtime": 0.7355, | |
| "eval_samples_per_second": 1418.114, | |
| "eval_steps_per_second": 89.737, | |
| "step": 2138 | |
| }, | |
| { | |
| "epoch": 1.1693171188026192, | |
| "grad_norm": 41.39296340942383, | |
| "learning_rate": 8.39714989228955e-06, | |
| "loss": 0.457, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.4031805425631432, | |
| "grad_norm": 0.10284680873155594, | |
| "learning_rate": 6.0330874226134376e-06, | |
| "loss": 0.403, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.637043966323667, | |
| "grad_norm": 15.211405754089355, | |
| "learning_rate": 3.6690249529373256e-06, | |
| "loss": 0.4216, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.8709073900841908, | |
| "grad_norm": 0.4435730576515198, | |
| "learning_rate": 1.304962483261214e-06, | |
| "loss": 0.408, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8785123229026794, | |
| "eval_matthews_correlation": 0.47051816449999595, | |
| "eval_runtime": 0.8593, | |
| "eval_samples_per_second": 1213.801, | |
| "eval_steps_per_second": 76.808, | |
| "step": 4276 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4276, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 65141619465252.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 2.021746224067011e-05, | |
| "num_train_epochs": 2, | |
| "per_device_train_batch_size": 4, | |
| "seed": 21 | |
| } | |
| } | |