|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"eval_steps": 500, |
|
"global_step": 21, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 10.54248783881401, |
|
"learning_rate": 1.6666666666666667e-06, |
|
"loss": 1.7384, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 6.430060136710083, |
|
"learning_rate": 4.849231551964771e-06, |
|
"loss": 1.7267, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 1.622734546661377, |
|
"eval_runtime": 1.5282, |
|
"eval_samples_per_second": 8.507, |
|
"eval_steps_per_second": 0.654, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"grad_norm": 6.294394428589208, |
|
"learning_rate": 3.3550503583141726e-06, |
|
"loss": 1.484, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_loss": 1.6291110515594482, |
|
"eval_runtime": 1.3246, |
|
"eval_samples_per_second": 9.814, |
|
"eval_steps_per_second": 0.755, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 2.14, |
|
"grad_norm": 6.200852884671378, |
|
"learning_rate": 1.2500000000000007e-06, |
|
"loss": 1.2674, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 2.86, |
|
"grad_norm": 4.814820729271434, |
|
"learning_rate": 3.798061746947995e-08, |
|
"loss": 1.1076, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_loss": 1.6497966051101685, |
|
"eval_runtime": 1.3247, |
|
"eval_samples_per_second": 9.814, |
|
"eval_steps_per_second": 0.755, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 21, |
|
"total_flos": 4344627855360.0, |
|
"train_loss": 1.3771353960037231, |
|
"train_runtime": 362.2329, |
|
"train_samples_per_second": 3.619, |
|
"train_steps_per_second": 0.058 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 21, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 500, |
|
"total_flos": 4344627855360.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|