|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.994418604651163, |
|
"eval_steps": 10000, |
|
"global_step": 134, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 5.033292770385742, |
|
"learning_rate": 9.253731343283582e-05, |
|
"loss": 1.2697, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"grad_norm": 2.4899826049804688, |
|
"learning_rate": 8.507462686567164e-05, |
|
"loss": 0.3049, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 1.5415112972259521, |
|
"learning_rate": 7.761194029850747e-05, |
|
"loss": 0.2733, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 1.638880968093872, |
|
"learning_rate": 7.014925373134329e-05, |
|
"loss": 0.2333, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"grad_norm": 1.1440837383270264, |
|
"learning_rate": 6.268656716417911e-05, |
|
"loss": 0.1867, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 1.1239829063415527, |
|
"learning_rate": 5.5223880597014934e-05, |
|
"loss": 0.192, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.04, |
|
"grad_norm": 1.2518362998962402, |
|
"learning_rate": 4.7761194029850745e-05, |
|
"loss": 0.1704, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"grad_norm": 0.8198823928833008, |
|
"learning_rate": 4.029850746268657e-05, |
|
"loss": 0.0894, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"grad_norm": 0.8759316205978394, |
|
"learning_rate": 3.283582089552239e-05, |
|
"loss": 0.1051, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"grad_norm": 1.0815106630325317, |
|
"learning_rate": 2.537313432835821e-05, |
|
"loss": 0.0918, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"grad_norm": 0.6786580085754395, |
|
"learning_rate": 1.791044776119403e-05, |
|
"loss": 0.0828, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"grad_norm": 0.5591555833816528, |
|
"learning_rate": 1.0447761194029851e-05, |
|
"loss": 0.0796, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"grad_norm": 0.9764952659606934, |
|
"learning_rate": 2.9850746268656716e-06, |
|
"loss": 0.0783, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"step": 134, |
|
"total_flos": 708532919205888.0, |
|
"train_loss": 0.23782958757521502, |
|
"train_runtime": 525.475, |
|
"train_samples_per_second": 32.728, |
|
"train_steps_per_second": 0.255 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 134, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 708532919205888.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|