|
{ |
|
"best_metric": 0.409545749280308, |
|
"best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/checkpoint-1069", |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 1069, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.47, |
|
"grad_norm": 8.771077156066895, |
|
"learning_rate": 6.473083427439518e-06, |
|
"loss": 0.5665, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"grad_norm": 4.481599807739258, |
|
"learning_rate": 4.497172735302912e-06, |
|
"loss": 0.4944, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.516472339630127, |
|
"eval_matthews_correlation": 0.409545749280308, |
|
"eval_runtime": 0.3422, |
|
"eval_samples_per_second": 3047.653, |
|
"eval_steps_per_second": 192.852, |
|
"step": 1069 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2138, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 37492413425856.0, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": { |
|
"learning_rate": 8.448994119576122e-06, |
|
"num_train_epochs": 2, |
|
"per_device_train_batch_size": 8, |
|
"seed": 22 |
|
} |
|
} |
|
|