| { | |
| "best_metric": 2.258429765701294, | |
| "best_model_checkpoint": "hBERTv1_new_pretrain_w_init__stsb/checkpoint-45", | |
| "epoch": 6.0, | |
| "global_step": 270, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.9200000000000004e-05, | |
| "loss": 2.5056, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_combined_score": 0.09202265464839787, | |
| "eval_loss": 2.258429765701294, | |
| "eval_pearson": 0.09487611226675945, | |
| "eval_runtime": 2.7043, | |
| "eval_samples_per_second": 554.669, | |
| "eval_spearmanr": 0.08916919703003628, | |
| "eval_steps_per_second": 4.437, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 2.1254, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_combined_score": 0.12405025742756695, | |
| "eval_loss": 2.687140941619873, | |
| "eval_pearson": 0.12502182632057038, | |
| "eval_runtime": 2.7046, | |
| "eval_samples_per_second": 554.603, | |
| "eval_spearmanr": 0.12307868853456354, | |
| "eval_steps_per_second": 4.437, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 1.9839, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_combined_score": 0.1814935022868095, | |
| "eval_loss": 2.270876884460449, | |
| "eval_pearson": 0.17897326966837357, | |
| "eval_runtime": 2.7033, | |
| "eval_samples_per_second": 554.887, | |
| "eval_spearmanr": 0.18401373490524542, | |
| "eval_steps_per_second": 4.439, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 1.6299, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_combined_score": 0.2744015884808604, | |
| "eval_loss": 2.511533498764038, | |
| "eval_pearson": 0.26907288016147085, | |
| "eval_runtime": 2.7032, | |
| "eval_samples_per_second": 554.907, | |
| "eval_spearmanr": 0.27973029680024986, | |
| "eval_steps_per_second": 4.439, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.3155, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_combined_score": 0.34449981946918246, | |
| "eval_loss": 2.455472707748413, | |
| "eval_pearson": 0.34527302370602114, | |
| "eval_runtime": 2.7116, | |
| "eval_samples_per_second": 553.177, | |
| "eval_spearmanr": 0.34372661523234377, | |
| "eval_steps_per_second": 4.425, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.9686, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_combined_score": 0.44888582263274857, | |
| "eval_loss": 2.8004186153411865, | |
| "eval_pearson": 0.4571367965831787, | |
| "eval_runtime": 2.7083, | |
| "eval_samples_per_second": 553.854, | |
| "eval_spearmanr": 0.4406348486823185, | |
| "eval_steps_per_second": 4.431, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "step": 270, | |
| "total_flos": 5108409405800448.0, | |
| "train_loss": 1.7548038623951099, | |
| "train_runtime": 260.2887, | |
| "train_samples_per_second": 1104.351, | |
| "train_steps_per_second": 8.644 | |
| } | |
| ], | |
| "max_steps": 2250, | |
| "num_train_epochs": 50, | |
| "total_flos": 5108409405800448.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |