| { | |
| "best_metric": 0.4863630663948402, | |
| "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/checkpoint-402", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 536, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.49098527431488037, | |
| "eval_matthews_correlation": 0.41038483457442687, | |
| "eval_runtime": 0.7284, | |
| "eval_samples_per_second": 1431.833, | |
| "eval_steps_per_second": 90.605, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5078911185264587, | |
| "eval_matthews_correlation": 0.4550840399220741, | |
| "eval_runtime": 1.0357, | |
| "eval_samples_per_second": 1007.09, | |
| "eval_steps_per_second": 63.728, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.6435816287994385, | |
| "eval_matthews_correlation": 0.4863630663948402, | |
| "eval_runtime": 0.7155, | |
| "eval_samples_per_second": 1457.781, | |
| "eval_steps_per_second": 92.247, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 3.7313432835820897, | |
| "grad_norm": 3.6653482913970947, | |
| "learning_rate": 5.49225906876445e-06, | |
| "loss": 0.2839, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.739212691783905, | |
| "eval_matthews_correlation": 0.4809785029448826, | |
| "eval_runtime": 0.8504, | |
| "eval_samples_per_second": 1226.417, | |
| "eval_steps_per_second": 77.606, | |
| "step": 536 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 536, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 210467669904852.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 8.177363502382625e-05, | |
| "num_train_epochs": 4, | |
| "per_device_train_batch_size": 64, | |
| "seed": 17 | |
| } | |
| } | |