|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 2.0, |
|
"eval_steps": 500, |
|
"global_step": 16740, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 1.9998805256869774e-05, |
|
"loss": 2.519, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"learning_rate": 1.850657108721625e-05, |
|
"loss": 0.4137, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 0.3, |
|
"learning_rate": 1.7013142174432496e-05, |
|
"loss": 0.2541, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 1.5519713261648747e-05, |
|
"loss": 0.2182, |
|
"step": 3750 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"learning_rate": 1.4026284348864996e-05, |
|
"loss": 0.1911, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 1.2532855436081244e-05, |
|
"loss": 0.1781, |
|
"step": 6250 |
|
}, |
|
{ |
|
"epoch": 0.9, |
|
"learning_rate": 1.1039426523297491e-05, |
|
"loss": 0.1595, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_ANIM": { |
|
"f1": 0.4893520616221114, |
|
"number": 1502, |
|
"precision": 0.3708791208791209, |
|
"recall": 0.7190412782956058 |
|
}, |
|
"eval_DIS": { |
|
"f1": 0.40930232558139534, |
|
"number": 740, |
|
"precision": 0.3120567375886525, |
|
"recall": 0.5945945945945946 |
|
}, |
|
"eval_LOC": { |
|
"f1": 0.9034779210629151, |
|
"number": 8492, |
|
"precision": 0.8589321728054347, |
|
"recall": 0.9528968440885539 |
|
}, |
|
"eval_ORG": { |
|
"f1": 0.7655367231638418, |
|
"number": 1328, |
|
"precision": 0.7207446808510638, |
|
"recall": 0.8162650602409639 |
|
}, |
|
"eval_PER": { |
|
"f1": 0.9023172029804307, |
|
"number": 5912, |
|
"precision": 0.874464370734804, |
|
"recall": 0.9320027063599459 |
|
}, |
|
"eval_loss": 0.17833726108074188, |
|
"eval_overall_accuracy": 0.9734603824262966, |
|
"eval_overall_f1": 0.8201001973584333, |
|
"eval_overall_precision": 0.7520883608687581, |
|
"eval_overall_recall": 0.9016356960053411, |
|
"eval_runtime": 132.2732, |
|
"eval_samples_per_second": 126.556, |
|
"eval_steps_per_second": 7.915, |
|
"step": 8370 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"learning_rate": 9.54599761051374e-06, |
|
"loss": 0.1378, |
|
"step": 8750 |
|
}, |
|
{ |
|
"epoch": 1.19, |
|
"learning_rate": 8.052568697729989e-06, |
|
"loss": 0.1172, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 1.34, |
|
"learning_rate": 6.5591397849462365e-06, |
|
"loss": 0.1087, |
|
"step": 11250 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 5.065710872162486e-06, |
|
"loss": 0.1069, |
|
"step": 12500 |
|
}, |
|
{ |
|
"epoch": 1.64, |
|
"learning_rate": 3.572281959378734e-06, |
|
"loss": 0.1024, |
|
"step": 13750 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"learning_rate": 2.078853046594982e-06, |
|
"loss": 0.0987, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 1.94, |
|
"learning_rate": 5.854241338112306e-07, |
|
"loss": 0.0939, |
|
"step": 16250 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_ANIM": { |
|
"f1": 0.5447112927950946, |
|
"number": 1502, |
|
"precision": 0.4419568822553897, |
|
"recall": 0.7097203728362184 |
|
}, |
|
"eval_DIS": { |
|
"f1": 0.46624472573839665, |
|
"number": 740, |
|
"precision": 0.38235294117647056, |
|
"recall": 0.5972972972972973 |
|
}, |
|
"eval_LOC": { |
|
"f1": 0.9167043911272069, |
|
"number": 8492, |
|
"precision": 0.8823529411764706, |
|
"recall": 0.9538389072067829 |
|
}, |
|
"eval_ORG": { |
|
"f1": 0.7596153846153847, |
|
"number": 1328, |
|
"precision": 0.6982323232323232, |
|
"recall": 0.8328313253012049 |
|
}, |
|
"eval_PER": { |
|
"f1": 0.9167771883289124, |
|
"number": 5912, |
|
"precision": 0.8988946684005201, |
|
"recall": 0.9353856562922869 |
|
}, |
|
"eval_loss": 0.19900059700012207, |
|
"eval_overall_accuracy": 0.9764447570932191, |
|
"eval_overall_f1": 0.8447657184460972, |
|
"eval_overall_precision": 0.7930091778949424, |
|
"eval_overall_recall": 0.9037498609102036, |
|
"eval_runtime": 131.794, |
|
"eval_samples_per_second": 127.016, |
|
"eval_steps_per_second": 7.944, |
|
"step": 16740 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"step": 16740, |
|
"total_flos": 1.5650156444757312e+16, |
|
"train_loss": 0.16563826671210669, |
|
"train_runtime": 3757.3365, |
|
"train_samples_per_second": 71.285, |
|
"train_steps_per_second": 4.455 |
|
} |
|
], |
|
"logging_steps": 1250, |
|
"max_steps": 16740, |
|
"num_train_epochs": 2, |
|
"save_steps": 500, |
|
"total_flos": 1.5650156444757312e+16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|