|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 9.433962264150944, |
|
"eval_steps": 500, |
|
"global_step": 3000, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.5803225806451613, |
|
"eval_loss": 0.19005867838859558, |
|
"eval_runtime": 5.674, |
|
"eval_samples_per_second": 546.354, |
|
"eval_steps_per_second": 11.456, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.5723270440251573, |
|
"grad_norm": 0.5149380564689636, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 0.3072, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.817741935483871, |
|
"eval_loss": 0.0924694836139679, |
|
"eval_runtime": 6.0294, |
|
"eval_samples_per_second": 514.15, |
|
"eval_steps_per_second": 10.781, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8770967741935484, |
|
"eval_loss": 0.06209348514676094, |
|
"eval_runtime": 5.7978, |
|
"eval_samples_per_second": 534.685, |
|
"eval_steps_per_second": 11.211, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.1446540880503147, |
|
"grad_norm": 0.4281909167766571, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.1072, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9009677419354839, |
|
"eval_loss": 0.048012129962444305, |
|
"eval_runtime": 5.7482, |
|
"eval_samples_per_second": 539.302, |
|
"eval_steps_per_second": 11.308, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.716981132075472, |
|
"grad_norm": 0.2933551073074341, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.0684, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9093548387096774, |
|
"eval_loss": 0.03922256454825401, |
|
"eval_runtime": 5.9235, |
|
"eval_samples_per_second": 523.338, |
|
"eval_steps_per_second": 10.973, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.912258064516129, |
|
"eval_loss": 0.03422800451517105, |
|
"eval_runtime": 5.7032, |
|
"eval_samples_per_second": 543.553, |
|
"eval_steps_per_second": 11.397, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 6.289308176100629, |
|
"grad_norm": 0.25101438164711, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.0534, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9209677419354839, |
|
"eval_loss": 0.0315035916864872, |
|
"eval_runtime": 5.8633, |
|
"eval_samples_per_second": 528.715, |
|
"eval_steps_per_second": 11.086, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.861635220125786, |
|
"grad_norm": 0.26583123207092285, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0462, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9264516129032258, |
|
"eval_loss": 0.029175369068980217, |
|
"eval_runtime": 5.7501, |
|
"eval_samples_per_second": 539.117, |
|
"eval_steps_per_second": 11.304, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9287096774193548, |
|
"eval_loss": 0.028132880106568336, |
|
"eval_runtime": 5.7642, |
|
"eval_samples_per_second": 537.802, |
|
"eval_steps_per_second": 11.276, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 9.433962264150944, |
|
"grad_norm": 0.24779267609119415, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0429, |
|
"step": 3000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 780738843279612.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.7637668053146042, |
|
"num_train_epochs": 10, |
|
"temperature": 18 |
|
} |
|
} |
|
|