| { | |
| "best_metric": 0.44018369913101196, | |
| "best_model_checkpoint": "hBERTv1_no_pretrain_qqp/checkpoint-11373", | |
| "epoch": 8.0, | |
| "global_step": 30328, | |
| "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": 0.5334, | |
| "step": 3791 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7675735839722978, | |
| "eval_combined_score": 0.7162763638068912, | |
| "eval_f1": 0.6649791436414846, | |
| "eval_loss": 0.4826056957244873, | |
| "eval_runtime": 188.5117, | |
| "eval_samples_per_second": 214.469, | |
| "eval_steps_per_second": 2.239, | |
| "step": 3791 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.8400000000000005e-05, | |
| "loss": 0.4491, | |
| "step": 7582 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7908731140242394, | |
| "eval_combined_score": 0.7417148741642232, | |
| "eval_f1": 0.6925566343042071, | |
| "eval_loss": 0.4492672383785248, | |
| "eval_runtime": 188.5255, | |
| "eval_samples_per_second": 214.454, | |
| "eval_steps_per_second": 2.238, | |
| "step": 7582 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.76e-05, | |
| "loss": 0.3866, | |
| "step": 11373 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7953747217412812, | |
| "eval_combined_score": 0.7611556910683499, | |
| "eval_f1": 0.7269366603954186, | |
| "eval_loss": 0.44018369913101196, | |
| "eval_runtime": 188.9583, | |
| "eval_samples_per_second": 213.963, | |
| "eval_steps_per_second": 2.233, | |
| "step": 11373 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.680000000000001e-05, | |
| "loss": 0.3657, | |
| "step": 15164 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.777491961414791, | |
| "eval_combined_score": 0.7493077018308378, | |
| "eval_f1": 0.7211234422468845, | |
| "eval_loss": 0.4990079700946808, | |
| "eval_runtime": 188.3561, | |
| "eval_samples_per_second": 214.647, | |
| "eval_steps_per_second": 2.24, | |
| "step": 15164 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.3708, | |
| "step": 18955 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8076675735839723, | |
| "eval_combined_score": 0.7674797151079011, | |
| "eval_f1": 0.72729185663183, | |
| "eval_loss": 0.474396288394928, | |
| "eval_runtime": 188.6326, | |
| "eval_samples_per_second": 214.332, | |
| "eval_steps_per_second": 2.237, | |
| "step": 18955 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.52e-05, | |
| "loss": 0.2948, | |
| "step": 22746 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8143210487261934, | |
| "eval_combined_score": 0.7760979872633377, | |
| "eval_f1": 0.737874925800482, | |
| "eval_loss": 0.4692937135696411, | |
| "eval_runtime": 189.2189, | |
| "eval_samples_per_second": 213.668, | |
| "eval_steps_per_second": 2.23, | |
| "step": 22746 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.44e-05, | |
| "loss": 0.2546, | |
| "step": 26537 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8119960425426663, | |
| "eval_combined_score": 0.7848819882556219, | |
| "eval_f1": 0.7577679339685777, | |
| "eval_loss": 0.45072534680366516, | |
| "eval_runtime": 189.1806, | |
| "eval_samples_per_second": 213.711, | |
| "eval_steps_per_second": 2.231, | |
| "step": 26537 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 0.2225, | |
| "step": 30328 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8193173386099432, | |
| "eval_combined_score": 0.7852080917725173, | |
| "eval_f1": 0.7510988449350915, | |
| "eval_loss": 0.5244665145874023, | |
| "eval_runtime": 189.2507, | |
| "eval_samples_per_second": 213.632, | |
| "eval_steps_per_second": 2.23, | |
| "step": 30328 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 30328, | |
| "total_flos": 4.310753534195794e+17, | |
| "train_loss": 0.3596843980101412, | |
| "train_runtime": 42048.5829, | |
| "train_samples_per_second": 432.65, | |
| "train_steps_per_second": 4.508 | |
| } | |
| ], | |
| "max_steps": 189550, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.310753534195794e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |