|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.8, |
|
"global_step": 15, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.0002, |
|
"loss": 1.2863, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 0.00019749279121818235, |
|
"loss": 1.2544, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 0.0001900968867902419, |
|
"loss": 1.0333, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"learning_rate": 0.000178183148246803, |
|
"loss": 0.9202, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"learning_rate": 0.00016234898018587337, |
|
"loss": 0.8506, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.92, |
|
"learning_rate": 0.00014338837391175582, |
|
"loss": 0.7345, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 2.24, |
|
"learning_rate": 0.00012225209339563145, |
|
"loss": 0.6905, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 2.56, |
|
"learning_rate": 0.0001, |
|
"loss": 0.6544, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 2.88, |
|
"learning_rate": 7.774790660436858e-05, |
|
"loss": 0.6118, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 3.2, |
|
"learning_rate": 5.6611626088244194e-05, |
|
"loss": 0.5865, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 3.52, |
|
"learning_rate": 3.7651019814126654e-05, |
|
"loss": 0.5699, |
|
"step": 11 |
|
}, |
|
{ |
|
"epoch": 3.84, |
|
"learning_rate": 2.181685175319702e-05, |
|
"loss": 0.505, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 4.16, |
|
"learning_rate": 9.903113209758096e-06, |
|
"loss": 0.4942, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 4.48, |
|
"learning_rate": 2.5072087818176382e-06, |
|
"loss": 0.5002, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"learning_rate": 0.0, |
|
"loss": 0.4813, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 4.8, |
|
"step": 15, |
|
"total_flos": 4296337784832.0, |
|
"train_loss": 0.7448747992515564, |
|
"train_runtime": 1150.9521, |
|
"train_samples_per_second": 0.869, |
|
"train_steps_per_second": 0.013 |
|
} |
|
], |
|
"max_steps": 15, |
|
"num_train_epochs": 5, |
|
"total_flos": 4296337784832.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|