|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.9948253557567917, |
|
"eval_steps": 500, |
|
"global_step": 2316, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.8260741233825684, |
|
"learning_rate": 0.0001501296456352636, |
|
"loss": 0.9352, |
|
"step": 579 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.8111525177955627, |
|
"eval_runtime": 7.5351, |
|
"eval_samples_per_second": 6.636, |
|
"eval_steps_per_second": 1.725, |
|
"step": 579 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 0.8390282988548279, |
|
"learning_rate": 0.0001, |
|
"loss": 0.7267, |
|
"step": 1159 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 0.7883608937263489, |
|
"eval_runtime": 7.5372, |
|
"eval_samples_per_second": 6.634, |
|
"eval_steps_per_second": 1.725, |
|
"step": 1159 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 0.9434927105903625, |
|
"learning_rate": 4.987035436473639e-05, |
|
"loss": 0.619, |
|
"step": 1739 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 0.8036102056503296, |
|
"eval_runtime": 7.5564, |
|
"eval_samples_per_second": 6.617, |
|
"eval_steps_per_second": 1.72, |
|
"step": 1739 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"grad_norm": 1.1188745498657227, |
|
"learning_rate": 0.0, |
|
"loss": 0.5011, |
|
"step": 2316 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"eval_loss": 0.8656747341156006, |
|
"eval_runtime": 7.5297, |
|
"eval_samples_per_second": 6.64, |
|
"eval_steps_per_second": 1.726, |
|
"step": 2316 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2316, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"total_flos": 1.2227748710377882e+17, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|