|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.0, |
|
"global_step": 4665, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 2.6784565916398713e-05, |
|
"loss": 2.1905, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"eval_exact_match": 61.28666035950804, |
|
"eval_f1": 71.363458978244, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"learning_rate": 2.356913183279743e-05, |
|
"loss": 1.4634, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.64, |
|
"eval_exact_match": 67.86187322611164, |
|
"eval_f1": 77.44564129629615, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"learning_rate": 2.0353697749196142e-05, |
|
"loss": 1.2925, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 0.96, |
|
"eval_exact_match": 70.5771050141911, |
|
"eval_f1": 79.53930047182527, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"learning_rate": 1.7138263665594855e-05, |
|
"loss": 1.0802, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.29, |
|
"eval_exact_match": 72.1286660359508, |
|
"eval_f1": 80.97163819236762, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 1.392282958199357e-05, |
|
"loss": 1.0438, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"eval_exact_match": 73.28287606433302, |
|
"eval_f1": 81.68746817914159, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"learning_rate": 1.0707395498392284e-05, |
|
"loss": 1.0276, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 1.93, |
|
"eval_exact_match": 72.99905392620624, |
|
"eval_f1": 81.97780926342594, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"learning_rate": 7.491961414790997e-06, |
|
"loss": 0.8789, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 2.25, |
|
"eval_exact_match": 74.03973509933775, |
|
"eval_f1": 82.60781680674411, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"learning_rate": 4.276527331189711e-06, |
|
"loss": 0.8502, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"eval_exact_match": 74.33301797540209, |
|
"eval_f1": 82.95704498318004, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"learning_rate": 1.0610932475884245e-06, |
|
"loss": 0.8325, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"eval_exact_match": 74.5600756859035, |
|
"eval_f1": 83.17536432572425, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 4665, |
|
"total_flos": 5.848017388319232e+16, |
|
"train_loss": 1.1721088012826149, |
|
"train_runtime": 6793.9155, |
|
"train_samples_per_second": 43.923, |
|
"train_steps_per_second": 0.687 |
|
} |
|
], |
|
"max_steps": 4665, |
|
"num_train_epochs": 3, |
|
"total_flos": 5.848017388319232e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|