|
{ |
|
"best_metric": 2.252106189727783, |
|
"best_model_checkpoint": "add_BERT_no_pretrain_stsb/checkpoint-270", |
|
"epoch": 11.0, |
|
"global_step": 495, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 0.0004915555555555556, |
|
"loss": 6.3185, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_combined_score": -0.0010166901927902152, |
|
"eval_loss": 2.764800786972046, |
|
"eval_pearson": 0.0033488460201501727, |
|
"eval_runtime": 2.3463, |
|
"eval_samples_per_second": 639.317, |
|
"eval_spearmanr": -0.005382226405730603, |
|
"eval_steps_per_second": 5.115, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 0.0004815555555555556, |
|
"loss": 2.2915, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_combined_score": 0.015394997836546201, |
|
"eval_loss": 2.2692453861236572, |
|
"eval_pearson": 0.020749248741372028, |
|
"eval_runtime": 2.337, |
|
"eval_samples_per_second": 641.849, |
|
"eval_spearmanr": 0.010040746931720372, |
|
"eval_steps_per_second": 5.135, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 0.0004715555555555556, |
|
"loss": 2.2747, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_combined_score": 0.010336013790419181, |
|
"eval_loss": 2.362309217453003, |
|
"eval_pearson": 0.016689666809067433, |
|
"eval_runtime": 2.3411, |
|
"eval_samples_per_second": 640.734, |
|
"eval_spearmanr": 0.003982360771770928, |
|
"eval_steps_per_second": 5.126, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 0.00046155555555555555, |
|
"loss": 2.2372, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_combined_score": 0.006720541376488365, |
|
"eval_loss": 2.883573293685913, |
|
"eval_pearson": 0.009008075092391842, |
|
"eval_runtime": 2.342, |
|
"eval_samples_per_second": 640.473, |
|
"eval_spearmanr": 0.004433007660584889, |
|
"eval_steps_per_second": 5.124, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"learning_rate": 0.0004515555555555556, |
|
"loss": 2.2573, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_combined_score": 0.035108956445297795, |
|
"eval_loss": 2.2528281211853027, |
|
"eval_pearson": 0.03370956593774324, |
|
"eval_runtime": 2.3442, |
|
"eval_samples_per_second": 639.885, |
|
"eval_spearmanr": 0.03650834695285234, |
|
"eval_steps_per_second": 5.119, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"learning_rate": 0.00044155555555555555, |
|
"loss": 2.1979, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_combined_score": 0.016057292835545172, |
|
"eval_loss": 2.252106189727783, |
|
"eval_pearson": 0.014231574810207417, |
|
"eval_runtime": 2.3413, |
|
"eval_samples_per_second": 640.679, |
|
"eval_spearmanr": 0.017883010860882925, |
|
"eval_steps_per_second": 5.125, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"learning_rate": 0.0004315555555555556, |
|
"loss": 2.2244, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_combined_score": 0.01728166633294926, |
|
"eval_loss": 2.3162295818328857, |
|
"eval_pearson": 0.015679956373000268, |
|
"eval_runtime": 2.3435, |
|
"eval_samples_per_second": 640.074, |
|
"eval_spearmanr": 0.018883376292898248, |
|
"eval_steps_per_second": 5.121, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"learning_rate": 0.0004215555555555556, |
|
"loss": 2.1832, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_combined_score": 0.002281605635553319, |
|
"eval_loss": 2.3739013671875, |
|
"eval_pearson": 0.0006456617990473748, |
|
"eval_runtime": 2.3491, |
|
"eval_samples_per_second": 638.533, |
|
"eval_spearmanr": 0.003917549472059263, |
|
"eval_steps_per_second": 5.108, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"learning_rate": 0.0004115555555555556, |
|
"loss": 2.3249, |
|
"step": 405 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_combined_score": NaN, |
|
"eval_loss": 2.3828952312469482, |
|
"eval_pearson": NaN, |
|
"eval_runtime": 2.3478, |
|
"eval_samples_per_second": 638.891, |
|
"eval_spearmanr": NaN, |
|
"eval_steps_per_second": 5.111, |
|
"step": 405 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"learning_rate": 0.00040155555555555555, |
|
"loss": 2.1956, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_combined_score": NaN, |
|
"eval_loss": 2.3083081245422363, |
|
"eval_pearson": NaN, |
|
"eval_runtime": 2.3435, |
|
"eval_samples_per_second": 640.08, |
|
"eval_spearmanr": NaN, |
|
"eval_steps_per_second": 5.121, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"learning_rate": 0.0003915555555555556, |
|
"loss": 2.2148, |
|
"step": 495 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_combined_score": -0.027554660458158586, |
|
"eval_loss": 2.2706003189086914, |
|
"eval_pearson": -0.028324243354752993, |
|
"eval_runtime": 2.361, |
|
"eval_samples_per_second": 635.329, |
|
"eval_spearmanr": -0.026785077561564178, |
|
"eval_steps_per_second": 5.083, |
|
"step": 495 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"step": 495, |
|
"total_flos": 9031970478620672.0, |
|
"train_loss": 2.6108932957504734, |
|
"train_runtime": 375.1993, |
|
"train_samples_per_second": 766.126, |
|
"train_steps_per_second": 5.997 |
|
} |
|
], |
|
"max_steps": 2250, |
|
"num_train_epochs": 50, |
|
"total_flos": 9031970478620672.0, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|