|
{ |
|
"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": 10.339241095395584, |
|
"learning_rate": 1.6666666666666667e-06, |
|
"loss": 1.7184, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 6.5447987894645525, |
|
"learning_rate": 4.849231551964771e-06, |
|
"loss": 1.7037, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 1.7993109226226807, |
|
"eval_runtime": 1.5526, |
|
"eval_samples_per_second": 5.797, |
|
"eval_steps_per_second": 0.644, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 5.9647734842541436, |
|
"learning_rate": 3.3550503583141726e-06, |
|
"loss": 1.5059, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 1.8139997720718384, |
|
"eval_runtime": 1.328, |
|
"eval_samples_per_second": 6.777, |
|
"eval_steps_per_second": 0.753, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 6.307006600680606, |
|
"learning_rate": 1.2500000000000007e-06, |
|
"loss": 1.2827, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 4.578735827580534, |
|
"learning_rate": 3.798061746947995e-08, |
|
"loss": 1.0945, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.8357751369476318, |
|
"eval_runtime": 1.327, |
|
"eval_samples_per_second": 6.782, |
|
"eval_steps_per_second": 0.754, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 21, |
|
"total_flos": 4344627855360.0, |
|
"train_loss": 1.3847338415327526, |
|
"train_runtime": 362.3075, |
|
"train_samples_per_second": 3.66, |
|
"train_steps_per_second": 0.058 |
|
} |
|
], |
|
"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 |
|
} |
|
|