|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.0007662394881520219, |
|
"eval_steps": 3, |
|
"global_step": 10, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 7.662394881520219e-05, |
|
"grad_norm": 2.0827770233154297, |
|
"learning_rate": 2e-05, |
|
"loss": 0.8855, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 7.662394881520219e-05, |
|
"eval_loss": 0.8962417840957642, |
|
"eval_runtime": 1007.1641, |
|
"eval_samples_per_second": 5.456, |
|
"eval_steps_per_second": 2.728, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.00015324789763040438, |
|
"grad_norm": 2.42187237739563, |
|
"learning_rate": 4e-05, |
|
"loss": 0.8898, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.00022987184644560657, |
|
"grad_norm": 2.237582206726074, |
|
"learning_rate": 6e-05, |
|
"loss": 0.8603, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.00022987184644560657, |
|
"eval_loss": 0.8306677937507629, |
|
"eval_runtime": 942.596, |
|
"eval_samples_per_second": 5.83, |
|
"eval_steps_per_second": 2.915, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.00030649579526080876, |
|
"grad_norm": 2.084083318710327, |
|
"learning_rate": 8e-05, |
|
"loss": 0.8819, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.00038311974407601095, |
|
"grad_norm": 1.8943278789520264, |
|
"learning_rate": 0.0001, |
|
"loss": 0.6412, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.00045974369289121313, |
|
"grad_norm": 1.8883026838302612, |
|
"learning_rate": 0.00012, |
|
"loss": 0.4795, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.00045974369289121313, |
|
"eval_loss": 0.3448009192943573, |
|
"eval_runtime": 950.9924, |
|
"eval_samples_per_second": 5.778, |
|
"eval_steps_per_second": 2.89, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.0005363676417064154, |
|
"grad_norm": 1.984823226928711, |
|
"learning_rate": 0.00014, |
|
"loss": 0.3711, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.0006129915905216175, |
|
"grad_norm": 1.396019458770752, |
|
"learning_rate": 0.00016, |
|
"loss": 0.2359, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 0.0006896155393368198, |
|
"grad_norm": 1.4580814838409424, |
|
"learning_rate": 0.00018, |
|
"loss": 0.1361, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.0006896155393368198, |
|
"eval_loss": 0.08926785737276077, |
|
"eval_runtime": 935.1615, |
|
"eval_samples_per_second": 5.876, |
|
"eval_steps_per_second": 2.939, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 0.0007662394881520219, |
|
"grad_norm": 0.6949341297149658, |
|
"learning_rate": 0.0002, |
|
"loss": 0.0878, |
|
"step": 10 |
|
} |
|
], |
|
"logging_steps": 1, |
|
"max_steps": 10, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 3, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3292692008140800.0, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|