|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"global_step": 324, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 3e-05, |
|
"loss": 1.1447, |
|
"step": 108 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_exact_match": 72.42625626675994, |
|
"eval_f1": 75.81685955023579, |
|
"eval_runtime": 3.7493, |
|
"eval_samples_per_second": 2287.598, |
|
"eval_steps_per_second": 7.201, |
|
"step": 108 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 3e-05, |
|
"loss": 0.4653, |
|
"step": 216 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_exact_match": 73.17243791535502, |
|
"eval_f1": 76.36095249868008, |
|
"eval_runtime": 3.0711, |
|
"eval_samples_per_second": 2792.849, |
|
"eval_steps_per_second": 8.792, |
|
"step": 216 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 3e-05, |
|
"loss": 0.4212, |
|
"step": 324 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_exact_match": 72.76436982627959, |
|
"eval_f1": 76.4819755764785, |
|
"eval_runtime": 3.1031, |
|
"eval_samples_per_second": 2763.993, |
|
"eval_steps_per_second": 8.701, |
|
"step": 324 |
|
} |
|
], |
|
"max_steps": 1080, |
|
"num_train_epochs": 10, |
|
"total_flos": 38745933873152.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|