|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.0, |
|
"eval_steps": 500, |
|
"global_step": 2738, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 8.173849525200878e-05, |
|
"loss": 1.8198, |
|
"step": 500 |
|
}, |
|
{ |
|
"None_loss": 1.661426305770874, |
|
"None_runtime": 129.2092, |
|
"None_samples_per_second": 115.665, |
|
"None_steps_per_second": 14.465, |
|
"epoch": 0.37, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 6.347699050401752e-05, |
|
"loss": 1.5114, |
|
"step": 1000 |
|
}, |
|
{ |
|
"None_loss": 1.5952973365783691, |
|
"None_runtime": 129.3057, |
|
"None_samples_per_second": 115.579, |
|
"None_steps_per_second": 14.454, |
|
"epoch": 0.73, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.1, |
|
"learning_rate": 4.52154857560263e-05, |
|
"loss": 1.4445, |
|
"step": 1500 |
|
}, |
|
{ |
|
"None_loss": 1.5740190744400024, |
|
"None_runtime": 129.646, |
|
"None_samples_per_second": 115.275, |
|
"None_steps_per_second": 14.416, |
|
"epoch": 1.1, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.46, |
|
"learning_rate": 2.695398100803506e-05, |
|
"loss": 1.4118, |
|
"step": 2000 |
|
}, |
|
{ |
|
"None_loss": 1.5628818273544312, |
|
"None_runtime": 129.6751, |
|
"None_samples_per_second": 115.25, |
|
"None_steps_per_second": 14.413, |
|
"epoch": 1.46, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.83, |
|
"learning_rate": 8.692476260043827e-06, |
|
"loss": 1.3941, |
|
"step": 2500 |
|
}, |
|
{ |
|
"None_loss": 1.5557925701141357, |
|
"None_runtime": 128.5071, |
|
"None_samples_per_second": 116.297, |
|
"None_steps_per_second": 14.544, |
|
"epoch": 1.83, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"step": 2738, |
|
"total_flos": 1.7783709700718592e+16, |
|
"train_loss": 1.505635230187694, |
|
"train_runtime": 4543.9022, |
|
"train_samples_per_second": 38.557, |
|
"train_steps_per_second": 0.603 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2738, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 1.7783709700718592e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|