|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 21, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 14.063593700055563, |
|
"learning_rate": 6.666666666666667e-06, |
|
"loss": 1.7596, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 6.212270277659958, |
|
"learning_rate": 1.9396926207859085e-05, |
|
"loss": 1.5447, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 1.628584623336792, |
|
"eval_runtime": 1.523, |
|
"eval_samples_per_second": 5.253, |
|
"eval_steps_per_second": 0.657, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 1.1593776542480763, |
|
"learning_rate": 1.342020143325669e-05, |
|
"loss": 1.2678, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 1.585951328277588, |
|
"eval_runtime": 1.5683, |
|
"eval_samples_per_second": 5.101, |
|
"eval_steps_per_second": 0.638, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 0.7675988471758454, |
|
"learning_rate": 5.000000000000003e-06, |
|
"loss": 1.1385, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 0.6673940691818862, |
|
"learning_rate": 1.519224698779198e-07, |
|
"loss": 1.0816, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.5817556381225586, |
|
"eval_runtime": 1.5965, |
|
"eval_samples_per_second": 5.011, |
|
"eval_steps_per_second": 0.626, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 21, |
|
"total_flos": 4344627855360.0, |
|
"train_loss": 1.2594334341230846, |
|
"train_runtime": 415.8691, |
|
"train_samples_per_second": 3.138, |
|
"train_steps_per_second": 0.05 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 21, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 4344627855360.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|