|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.716981132075472, |
|
"eval_steps": 500, |
|
"global_step": 1500, |
|
"is_hyper_param_search": true, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.6735483870967742, |
|
"eval_loss": 0.40811532735824585, |
|
"eval_runtime": 1.4302, |
|
"eval_samples_per_second": 2167.455, |
|
"eval_steps_per_second": 45.447, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 1.650593990216632e-05, |
|
"loss": 0.6421, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8425806451612903, |
|
"eval_loss": 0.14079666137695312, |
|
"eval_runtime": 1.4492, |
|
"eval_samples_per_second": 2139.121, |
|
"eval_steps_per_second": 44.853, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8938709677419355, |
|
"eval_loss": 0.07184918969869614, |
|
"eval_runtime": 1.4642, |
|
"eval_samples_per_second": 2117.238, |
|
"eval_steps_per_second": 44.394, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 1.3011879804332637e-05, |
|
"loss": 0.1641, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9141935483870968, |
|
"eval_loss": 0.05170894414186478, |
|
"eval_runtime": 1.4545, |
|
"eval_samples_per_second": 2131.259, |
|
"eval_steps_per_second": 44.688, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 9.517819706498952e-06, |
|
"loss": 0.0819, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 2862, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 9, |
|
"save_steps": 500, |
|
"total_flos": 391111083046572.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.749953517627743, |
|
"num_train_epochs": 9, |
|
"temperature": 2 |
|
} |
|
} |
|
|