|
{ |
|
"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": 0.7290322580645161, |
|
"eval_loss": 3.165098190307617, |
|
"eval_runtime": 2.394, |
|
"eval_samples_per_second": 1294.905, |
|
"eval_steps_per_second": 27.151, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 1.685534591194969e-05, |
|
"loss": 3.6948, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.86, |
|
"eval_loss": 1.5545812845230103, |
|
"eval_runtime": 2.45, |
|
"eval_samples_per_second": 1265.306, |
|
"eval_steps_per_second": 26.531, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9132258064516129, |
|
"eval_loss": 0.7676715850830078, |
|
"eval_runtime": 2.444, |
|
"eval_samples_per_second": 1268.413, |
|
"eval_steps_per_second": 26.596, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 1.371069182389937e-05, |
|
"loss": 1.3265, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9354838709677419, |
|
"eval_loss": 0.45812979340553284, |
|
"eval_runtime": 2.443, |
|
"eval_samples_per_second": 1268.93, |
|
"eval_steps_per_second": 26.607, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.4291, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 496057080049020.0, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.3328723302231085, |
|
"num_train_epochs": 10, |
|
"temperature": 5 |
|
} |
|
} |
|
|