|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 7.861635220125786, |
|
"eval_steps": 500, |
|
"global_step": 2500, |
|
"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": 2.385, |
|
"eval_samples_per_second": 1299.79, |
|
"eval_steps_per_second": 27.254, |
|
"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": 2.436, |
|
"eval_samples_per_second": 1272.579, |
|
"eval_steps_per_second": 26.683, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9132258064516129, |
|
"eval_loss": 0.7676715850830078, |
|
"eval_runtime": 2.407, |
|
"eval_samples_per_second": 1287.909, |
|
"eval_steps_per_second": 27.005, |
|
"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": 2.385, |
|
"eval_samples_per_second": 1299.792, |
|
"eval_steps_per_second": 27.254, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 1.0566037735849058e-05, |
|
"loss": 0.4291, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9438709677419355, |
|
"eval_loss": 0.3298114240169525, |
|
"eval_runtime": 2.445, |
|
"eval_samples_per_second": 1267.894, |
|
"eval_steps_per_second": 26.585, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9464516129032258, |
|
"eval_loss": 0.27686071395874023, |
|
"eval_runtime": 2.398, |
|
"eval_samples_per_second": 1292.742, |
|
"eval_steps_per_second": 27.106, |
|
"step": 1908 |
|
}, |
|
{ |
|
"epoch": 6.29, |
|
"learning_rate": 7.421383647798742e-06, |
|
"loss": 0.1725, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.947741935483871, |
|
"eval_loss": 0.25533074140548706, |
|
"eval_runtime": 2.452, |
|
"eval_samples_per_second": 1264.275, |
|
"eval_steps_per_second": 26.509, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.86, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0911, |
|
"step": 2500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 649636246191744.0, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.5311549230753155, |
|
"num_train_epochs": 10, |
|
"temperature": 2 |
|
} |
|
} |
|
|