|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.6, |
|
"eval_steps": 200, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 8.260869565217392e-06, |
|
"loss": 0.693, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_loss": 0.6899820566177368, |
|
"eval_runtime": 1.6725, |
|
"eval_samples_per_second": 597.893, |
|
"eval_steps_per_second": 37.667, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 5.652173913043479e-06, |
|
"loss": 0.6843, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"eval_loss": 0.6105484366416931, |
|
"eval_runtime": 1.7064, |
|
"eval_samples_per_second": 586.021, |
|
"eval_steps_per_second": 36.919, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"learning_rate": 3.043478260869566e-06, |
|
"loss": 0.6336, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 4.347826086956522e-07, |
|
"loss": 0.5981, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"eval_loss": 0.5873401165008545, |
|
"eval_runtime": 1.7375, |
|
"eval_samples_per_second": 575.544, |
|
"eval_steps_per_second": 36.259, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 1.2, |
|
"learning_rate": 0.0, |
|
"loss": 0.5851, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 1.28, |
|
"eval_loss": 0.5873731970787048, |
|
"eval_runtime": 1.7565, |
|
"eval_samples_per_second": 569.31, |
|
"eval_steps_per_second": 35.867, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.44, |
|
"learning_rate": 0.0, |
|
"loss": 0.5863, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"eval_loss": 0.5873731970787048, |
|
"eval_runtime": 1.7602, |
|
"eval_samples_per_second": 568.102, |
|
"eval_steps_per_second": 35.79, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.6, |
|
"step": 1000, |
|
"total_flos": 364154381869824.0, |
|
"train_loss": 0.6259692611694336, |
|
"train_runtime": 219.9954, |
|
"train_samples_per_second": 72.729, |
|
"train_steps_per_second": 4.546 |
|
} |
|
], |
|
"logging_steps": 150, |
|
"max_steps": 1000, |
|
"num_train_epochs": 2, |
|
"save_steps": 200, |
|
"total_flos": 364154381869824.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|