| { | |
| "best_metric": 0.4221435487270355, | |
| "best_model_checkpoint": "./results_train/roberta-base/stsb/checkpoint-3000", | |
| "epoch": 10.0, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 1.832151300236407e-05, | |
| "loss": 1.6552, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_combined_score": 0.8925236922939852, | |
| "eval_loss": 0.5264632701873779, | |
| "eval_pearson": 0.8925146697908521, | |
| "eval_runtime": 4.2272, | |
| "eval_samples_per_second": 354.842, | |
| "eval_spearmanr": 0.8925327147971182, | |
| "eval_steps_per_second": 44.474, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.536643026004728e-05, | |
| "loss": 0.3579, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_combined_score": 0.9006678737191953, | |
| "eval_loss": 0.4626253545284271, | |
| "eval_pearson": 0.9021887771929038, | |
| "eval_runtime": 4.2213, | |
| "eval_samples_per_second": 355.343, | |
| "eval_spearmanr": 0.8991469702454868, | |
| "eval_steps_per_second": 44.536, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.2411347517730496e-05, | |
| "loss": 0.2198, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_combined_score": 0.904776169841008, | |
| "eval_loss": 0.4396342635154724, | |
| "eval_pearson": 0.9053811595680402, | |
| "eval_runtime": 4.231, | |
| "eval_samples_per_second": 354.53, | |
| "eval_spearmanr": 0.9041711801139757, | |
| "eval_steps_per_second": 44.434, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 9.456264775413712e-06, | |
| "loss": 0.1585, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_combined_score": 0.9060185098833846, | |
| "eval_loss": 0.45370450615882874, | |
| "eval_pearson": 0.9068592393148944, | |
| "eval_runtime": 4.2301, | |
| "eval_samples_per_second": 354.603, | |
| "eval_spearmanr": 0.9051777804518748, | |
| "eval_steps_per_second": 44.444, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 6.501182033096928e-06, | |
| "loss": 0.1139, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "eval_combined_score": 0.9078003245360166, | |
| "eval_loss": 0.49749574065208435, | |
| "eval_pearson": 0.909052291163071, | |
| "eval_runtime": 4.2316, | |
| "eval_samples_per_second": 354.473, | |
| "eval_spearmanr": 0.9065483579089622, | |
| "eval_steps_per_second": 44.427, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3.5460992907801423e-06, | |
| "loss": 0.0868, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_combined_score": 0.9104012240144247, | |
| "eval_loss": 0.4221435487270355, | |
| "eval_pearson": 0.911586582943305, | |
| "eval_runtime": 4.2435, | |
| "eval_samples_per_second": 353.484, | |
| "eval_spearmanr": 0.9092158650855444, | |
| "eval_steps_per_second": 44.303, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 5.91016548463357e-07, | |
| "loss": 0.073, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "eval_combined_score": 0.9086096499017097, | |
| "eval_loss": 0.4311141073703766, | |
| "eval_pearson": 0.9095606688872401, | |
| "eval_runtime": 4.2319, | |
| "eval_samples_per_second": 354.451, | |
| "eval_spearmanr": 0.9076586309161793, | |
| "eval_steps_per_second": 44.424, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 3600, | |
| "total_flos": 3781529690027520.0, | |
| "train_loss": 0.3720528290006849, | |
| "train_runtime": 717.805, | |
| "train_samples_per_second": 80.091, | |
| "train_steps_per_second": 5.015 | |
| } | |
| ], | |
| "max_steps": 3600, | |
| "num_train_epochs": 10, | |
| "total_flos": 3781529690027520.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |