|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 10.0, |
|
"eval_steps": 500, |
|
"global_step": 3180, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.594516129032258, |
|
"eval_loss": 0.20276567339897156, |
|
"eval_runtime": 5.2979, |
|
"eval_samples_per_second": 585.143, |
|
"eval_steps_per_second": 12.269, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.5723270440251573, |
|
"grad_norm": 0.535661518573761, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 0.3272, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8270967741935484, |
|
"eval_loss": 0.09640489518642426, |
|
"eval_runtime": 5.5004, |
|
"eval_samples_per_second": 563.593, |
|
"eval_steps_per_second": 11.817, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8841935483870967, |
|
"eval_loss": 0.06340694427490234, |
|
"eval_runtime": 5.3223, |
|
"eval_samples_per_second": 582.452, |
|
"eval_steps_per_second": 12.213, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.1446540880503147, |
|
"grad_norm": 0.4495888650417328, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 0.1121, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9006451612903226, |
|
"eval_loss": 0.048330824822187424, |
|
"eval_runtime": 5.6025, |
|
"eval_samples_per_second": 553.321, |
|
"eval_steps_per_second": 11.602, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.716981132075472, |
|
"grad_norm": 0.30502042174339294, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.0703, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9106451612903226, |
|
"eval_loss": 0.03915739059448242, |
|
"eval_runtime": 5.4572, |
|
"eval_samples_per_second": 568.056, |
|
"eval_steps_per_second": 11.911, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9151612903225806, |
|
"eval_loss": 0.03403216972947121, |
|
"eval_runtime": 5.2442, |
|
"eval_samples_per_second": 591.135, |
|
"eval_steps_per_second": 12.395, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 6.289308176100629, |
|
"grad_norm": 0.2584752142429352, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.0544, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9245161290322581, |
|
"eval_loss": 0.03124151937663555, |
|
"eval_runtime": 5.4453, |
|
"eval_samples_per_second": 569.297, |
|
"eval_steps_per_second": 11.937, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.861635220125786, |
|
"grad_norm": 0.2786354124546051, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.047, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9280645161290323, |
|
"eval_loss": 0.028918160125613213, |
|
"eval_runtime": 5.5666, |
|
"eval_samples_per_second": 556.888, |
|
"eval_steps_per_second": 11.677, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.93, |
|
"eval_loss": 0.027848755940794945, |
|
"eval_runtime": 5.4913, |
|
"eval_samples_per_second": 564.529, |
|
"eval_steps_per_second": 11.837, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 9.433962264150944, |
|
"grad_norm": 0.25698819756507874, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0435, |
|
"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": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 863743180573392.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.03960757682163807, |
|
"num_train_epochs": 10, |
|
"temperature": 9 |
|
} |
|
} |
|
|