|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.9595959595959593, |
|
"eval_steps": 500, |
|
"global_step": 196, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 2.071033477783203, |
|
"learning_rate": 0.00019545454545454548, |
|
"loss": 1.5121, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"grad_norm": 0.7647247314453125, |
|
"learning_rate": 0.00016704545454545455, |
|
"loss": 0.4934, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.52, |
|
"grad_norm": 0.6377527713775635, |
|
"learning_rate": 0.00013863636363636365, |
|
"loss": 0.2928, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"grad_norm": 1.0899125337600708, |
|
"learning_rate": 0.00011022727272727273, |
|
"loss": 0.1976, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"grad_norm": 0.8106021285057068, |
|
"learning_rate": 8.181818181818183e-05, |
|
"loss": 0.1196, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"grad_norm": 0.38758715987205505, |
|
"learning_rate": 5.340909090909091e-05, |
|
"loss": 0.1052, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 3.54, |
|
"grad_norm": 0.48243477940559387, |
|
"learning_rate": 2.5e-05, |
|
"loss": 0.0809, |
|
"step": 175 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 196, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"total_flos": 6.856828180758528e+16, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|