|
{ |
|
"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": 0.7290322580645161, |
|
"eval_loss": 3.165098190307617, |
|
"eval_runtime": 2.374, |
|
"eval_samples_per_second": 1305.814, |
|
"eval_steps_per_second": 27.38, |
|
"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": 2.435, |
|
"eval_samples_per_second": 1273.1, |
|
"eval_steps_per_second": 26.694, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9132258064516129, |
|
"eval_loss": 0.7676715850830078, |
|
"eval_runtime": 2.394, |
|
"eval_samples_per_second": 1294.904, |
|
"eval_steps_per_second": 27.151, |
|
"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": 2.44, |
|
"eval_samples_per_second": 1270.492, |
|
"eval_steps_per_second": 26.639, |
|
"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": 2.437, |
|
"eval_samples_per_second": 1272.054, |
|
"eval_steps_per_second": 26.672, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9464516129032258, |
|
"eval_loss": 0.27686071395874023, |
|
"eval_runtime": 2.402, |
|
"eval_samples_per_second": 1290.591, |
|
"eval_steps_per_second": 27.061, |
|
"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": 2.458, |
|
"eval_samples_per_second": 1261.188, |
|
"eval_steps_per_second": 26.444, |
|
"step": 2226 |
|
}, |
|
{ |
|
"epoch": 7.86, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0911, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.95, |
|
"eval_loss": 0.24502110481262207, |
|
"eval_runtime": 2.444, |
|
"eval_samples_per_second": 1268.412, |
|
"eval_steps_per_second": 26.596, |
|
"step": 2544 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9496774193548387, |
|
"eval_loss": 0.24202485382556915, |
|
"eval_runtime": 2.451, |
|
"eval_samples_per_second": 1264.79, |
|
"eval_steps_per_second": 26.52, |
|
"step": 2862 |
|
}, |
|
{ |
|
"epoch": 9.43, |
|
"learning_rate": 1.1320754716981133e-06, |
|
"loss": 0.0607, |
|
"step": 3000 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 3180, |
|
"num_train_epochs": 10, |
|
"save_steps": 500, |
|
"total_flos": 779832454697856.0, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.5743930191103429, |
|
"num_train_epochs": 10, |
|
"temperature": 7 |
|
} |
|
} |
|
|