|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.008128098837681867, |
|
"eval_steps": 50, |
|
"global_step": 50, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0001625619767536373, |
|
"eval_loss": 1.5123971700668335, |
|
"eval_runtime": 88.7745, |
|
"eval_samples_per_second": 29.186, |
|
"eval_steps_per_second": 14.599, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.0016256197675363732, |
|
"grad_norm": 1.664616584777832, |
|
"learning_rate": 0.0002, |
|
"loss": 5.9758, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.0032512395350727465, |
|
"grad_norm": 2.718796491622925, |
|
"learning_rate": 0.0002, |
|
"loss": 5.3996, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.00487685930260912, |
|
"grad_norm": 1.4594964981079102, |
|
"learning_rate": 0.0002, |
|
"loss": 5.1102, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.006502479070145493, |
|
"grad_norm": 2.452237367630005, |
|
"learning_rate": 0.0002, |
|
"loss": 4.8982, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.008128098837681867, |
|
"grad_norm": 1.6779563426971436, |
|
"learning_rate": 0.0002, |
|
"loss": 4.7516, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.008128098837681867, |
|
"eval_loss": 1.1936694383621216, |
|
"eval_runtime": 88.7928, |
|
"eval_samples_per_second": 29.18, |
|
"eval_steps_per_second": 14.596, |
|
"step": 50 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 200, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 50, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 8835964638068736.0, |
|
"train_batch_size": 2, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|