|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.988679245283019, |
|
"eval_steps": 100, |
|
"global_step": 198, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1509433962264151, |
|
"grad_norm": 1.5483485460281372, |
|
"learning_rate": 4e-06, |
|
"loss": 1.1432, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.3018867924528302, |
|
"grad_norm": 0.7621527314186096, |
|
"learning_rate": 8e-06, |
|
"loss": 0.9841, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.4528301886792453, |
|
"grad_norm": 0.5265641212463379, |
|
"learning_rate": 7.937861311446333e-06, |
|
"loss": 0.8367, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.6037735849056604, |
|
"grad_norm": 0.4887830317020416, |
|
"learning_rate": 7.753375854092917e-06, |
|
"loss": 0.7718, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.7547169811320755, |
|
"grad_norm": 0.3349996507167816, |
|
"learning_rate": 7.4522754701283365e-06, |
|
"loss": 0.745, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.9056603773584906, |
|
"grad_norm": 0.4875039756298065, |
|
"learning_rate": 7.04391515104387e-06, |
|
"loss": 0.6574, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0566037735849056, |
|
"grad_norm": 0.46130555868148804, |
|
"learning_rate": 6.540982384182154e-06, |
|
"loss": 0.6706, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.2075471698113207, |
|
"grad_norm": 0.45948392152786255, |
|
"learning_rate": 5.9591029608249135e-06, |
|
"loss": 0.6149, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.3584905660377358, |
|
"grad_norm": 0.4454590380191803, |
|
"learning_rate": 5.31635549310404e-06, |
|
"loss": 0.5374, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.509433962264151, |
|
"grad_norm": 0.42349863052368164, |
|
"learning_rate": 4.6327097233772155e-06, |
|
"loss": 0.5595, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.509433962264151, |
|
"eval_loss": 0.5344268083572388, |
|
"eval_runtime": 2.3441, |
|
"eval_samples_per_second": 47.779, |
|
"eval_steps_per_second": 2.986, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.6603773584905661, |
|
"grad_norm": 0.4312027096748352, |
|
"learning_rate": 3.929406077427486e-06, |
|
"loss": 0.5048, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.8113207547169812, |
|
"grad_norm": 0.5000870227813721, |
|
"learning_rate": 3.2282957383620122e-06, |
|
"loss": 0.5256, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.9622641509433962, |
|
"grad_norm": 0.46476179361343384, |
|
"learning_rate": 2.5511617446812707e-06, |
|
"loss": 0.5328, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 2.1132075471698113, |
|
"grad_norm": 0.36232122778892517, |
|
"learning_rate": 1.9190422055564713e-06, |
|
"loss": 0.4995, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 2.2641509433962264, |
|
"grad_norm": 0.3772554397583008, |
|
"learning_rate": 1.3515766605727983e-06, |
|
"loss": 0.4822, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 2.4150943396226414, |
|
"grad_norm": 0.4370742738246918, |
|
"learning_rate": 8.663958921125897e-07, |
|
"loss": 0.4785, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 2.5660377358490565, |
|
"grad_norm": 0.46978744864463806, |
|
"learning_rate": 4.785741485076355e-07, |
|
"loss": 0.4779, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.7169811320754715, |
|
"grad_norm": 0.4040665030479431, |
|
"learning_rate": 2.0016079702803678e-07, |
|
"loss": 0.5026, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.867924528301887, |
|
"grad_norm": 0.3927009701728821, |
|
"learning_rate": 3.9805957942182774e-08, |
|
"loss": 0.4935, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.988679245283019, |
|
"step": 198, |
|
"total_flos": 7.194515931031142e+16, |
|
"train_loss": 0.6248175543968124, |
|
"train_runtime": 424.9337, |
|
"train_samples_per_second": 14.953, |
|
"train_steps_per_second": 0.466 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 198, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 3, |
|
"save_steps": 200, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": true |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 7.194515931031142e+16, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|