|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.4827586206896552, |
|
"eval_steps": 500, |
|
"global_step": 72, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.3448275862068966, |
|
"grad_norm": 1.1789525747299194, |
|
"learning_rate": 0.00019618256431728194, |
|
"loss": 0.1525, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.6896551724137931, |
|
"grad_norm": 0.7358490824699402, |
|
"learning_rate": 0.00017390089172206592, |
|
"loss": 0.0224, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 1.0344827586206897, |
|
"grad_norm": 0.45523107051849365, |
|
"learning_rate": 0.00013612416661871533, |
|
"loss": 0.0108, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.3793103448275863, |
|
"grad_norm": 0.20521320402622223, |
|
"learning_rate": 9.077316405366981e-05, |
|
"loss": 0.0087, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.7241379310344827, |
|
"grad_norm": 0.22252900898456573, |
|
"learning_rate": 4.735678371226441e-05, |
|
"loss": 0.0086, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.0689655172413794, |
|
"grad_norm": 0.5247920155525208, |
|
"learning_rate": 1.4978286427038601e-05, |
|
"loss": 0.0078, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.413793103448276, |
|
"grad_norm": 0.3118768632411957, |
|
"learning_rate": 4.2658237049655323e-07, |
|
"loss": 0.0054, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.4827586206896552, |
|
"step": 72, |
|
"total_flos": 1425318441811200.0, |
|
"train_loss": 0.03014997382544809, |
|
"train_runtime": 44.0205, |
|
"train_samples_per_second": 26.17, |
|
"train_steps_per_second": 1.636 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 72, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 1425318441811200.0, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|