|
{ |
|
"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": 0.5735483870967742, |
|
"eval_loss": 0.19929757714271545, |
|
"eval_runtime": 1.4262, |
|
"eval_samples_per_second": 2173.603, |
|
"eval_steps_per_second": 45.576, |
|
"step": 318 |
|
}, |
|
{ |
|
"epoch": 1.57, |
|
"learning_rate": 1.4758909853249476e-05, |
|
"loss": 0.3159, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8096774193548387, |
|
"eval_loss": 0.10230626910924911, |
|
"eval_runtime": 1.4489, |
|
"eval_samples_per_second": 2139.6, |
|
"eval_steps_per_second": 44.863, |
|
"step": 636 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8670967741935484, |
|
"eval_loss": 0.0720420777797699, |
|
"eval_runtime": 1.4553, |
|
"eval_samples_per_second": 2130.203, |
|
"eval_steps_per_second": 44.666, |
|
"step": 954 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 9.517819706498952e-06, |
|
"loss": 0.1172, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.886774193548387, |
|
"eval_loss": 0.057750094681978226, |
|
"eval_runtime": 1.4432, |
|
"eval_samples_per_second": 2147.995, |
|
"eval_steps_per_second": 45.039, |
|
"step": 1272 |
|
}, |
|
{ |
|
"epoch": 4.72, |
|
"learning_rate": 4.276729559748428e-06, |
|
"loss": 0.0804, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 1908, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 6, |
|
"save_steps": 500, |
|
"total_flos": 391111083046572.0, |
|
"train_batch_size": 48, |
|
"trial_name": null, |
|
"trial_params": { |
|
"alpha": 0.3767704932875351, |
|
"num_train_epochs": 6, |
|
"temperature": 15 |
|
} |
|
} |
|
|