|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 0.4581061890146136, |
|
"global_step": 30000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 2.9999999999999997e-06, |
|
"loss": 0.8784, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 5.999999999999999e-06, |
|
"loss": 0.77, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 8.999999999999999e-06, |
|
"loss": 0.7664, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"learning_rate": 1.1999999999999999e-05, |
|
"loss": 0.7655, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"learning_rate": 1.4999999999999999e-05, |
|
"loss": 0.765, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"eval_runtime": 1.3797, |
|
"eval_samples_per_second": 724.791, |
|
"eval_steps_per_second": 11.597, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 1.7999999999999997e-05, |
|
"loss": 0.7647, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 2.1e-05, |
|
"loss": 0.7644, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"learning_rate": 2.3999999999999997e-05, |
|
"loss": 0.7638, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 2.6999999999999996e-05, |
|
"loss": 0.7633, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 2.9999999999999997e-05, |
|
"loss": 0.76, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"eval_runtime": 1.1376, |
|
"eval_samples_per_second": 879.066, |
|
"eval_steps_per_second": 14.065, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"learning_rate": 3.2999999999999996e-05, |
|
"loss": 0.7148, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 3.5999999999999994e-05, |
|
"loss": 0.6963, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 3.9e-05, |
|
"loss": 0.6755, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"learning_rate": 4.2e-05, |
|
"loss": 0.6516, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 4.4999999999999996e-05, |
|
"loss": 0.6412, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_runtime": 1.1689, |
|
"eval_samples_per_second": 855.472, |
|
"eval_steps_per_second": 13.688, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 0.24, |
|
"learning_rate": 4.7999999999999994e-05, |
|
"loss": 0.6348, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"learning_rate": 5.1e-05, |
|
"loss": 0.6295, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 5.399999999999999e-05, |
|
"loss": 0.6224, |
|
"step": 18000 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 5.6999999999999996e-05, |
|
"loss": 0.6169, |
|
"step": 19000 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"learning_rate": 5.9999999999999995e-05, |
|
"loss": 0.6113, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"eval_runtime": 1.0179, |
|
"eval_samples_per_second": 982.441, |
|
"eval_steps_per_second": 15.719, |
|
"step": 20000 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 6.299999999999999e-05, |
|
"loss": 0.6074, |
|
"step": 21000 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 6.599999999999999e-05, |
|
"loss": 0.6039, |
|
"step": 22000 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"learning_rate": 6.9e-05, |
|
"loss": 0.6005, |
|
"step": 23000 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"learning_rate": 7.199999999999999e-05, |
|
"loss": 0.5968, |
|
"step": 24000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"learning_rate": 7.5e-05, |
|
"loss": 0.5932, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"eval_runtime": 1.1249, |
|
"eval_samples_per_second": 888.989, |
|
"eval_steps_per_second": 14.224, |
|
"step": 25000 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"learning_rate": 7.8e-05, |
|
"loss": 0.5912, |
|
"step": 26000 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 8.1e-05, |
|
"loss": 0.58, |
|
"step": 27000 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 8.4e-05, |
|
"loss": 0.5698, |
|
"step": 28000 |
|
}, |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 8.699999999999999e-05, |
|
"loss": 0.5639, |
|
"step": 29000 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 8.999999999999999e-05, |
|
"loss": 0.5601, |
|
"step": 30000 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"eval_runtime": 1.0096, |
|
"eval_samples_per_second": 990.512, |
|
"eval_steps_per_second": 15.848, |
|
"step": 30000 |
|
} |
|
], |
|
"max_steps": 1000000, |
|
"num_train_epochs": 16, |
|
"total_flos": 2.1030078309104144e+21, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|