| { | |
| "best_metric": 0.8294392523364487, | |
| "best_model_checkpoint": "final_models/transformer_base_final_2/finetune/qnli/checkpoint-1600", | |
| "epoch": 9.836065573770492, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.09, | |
| "eval_accuracy": 0.7349081635475159, | |
| "eval_f1": 0.7757216876387861, | |
| "eval_loss": 0.53801429271698, | |
| "eval_mcc": 0.4737186740651392, | |
| "eval_runtime": 3.1305, | |
| "eval_samples_per_second": 730.228, | |
| "eval_steps_per_second": 91.358, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 4.316939890710383e-05, | |
| "loss": 0.6208, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_accuracy": 0.8057742714881897, | |
| "eval_f1": 0.824782951854775, | |
| "eval_loss": 0.42864909768104553, | |
| "eval_mcc": 0.6103254520818892, | |
| "eval_runtime": 3.1226, | |
| "eval_samples_per_second": 732.077, | |
| "eval_steps_per_second": 91.59, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.633879781420765e-05, | |
| "loss": 0.433, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "eval_accuracy": 0.8027121424674988, | |
| "eval_f1": 0.8223710122095312, | |
| "eval_loss": 0.46695178747177124, | |
| "eval_mcc": 0.604258498666583, | |
| "eval_runtime": 3.1147, | |
| "eval_samples_per_second": 733.949, | |
| "eval_steps_per_second": 91.824, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 2.9508196721311478e-05, | |
| "loss": 0.2954, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "eval_accuracy": 0.808398962020874, | |
| "eval_f1": 0.8294392523364487, | |
| "eval_loss": 0.5520442128181458, | |
| "eval_mcc": 0.6169307221841646, | |
| "eval_runtime": 3.1389, | |
| "eval_samples_per_second": 728.273, | |
| "eval_steps_per_second": 91.114, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.2677595628415303e-05, | |
| "loss": 0.1693, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "eval_accuracy": 0.7909011244773865, | |
| "eval_f1": 0.8138629283489095, | |
| "eval_loss": 0.7586725354194641, | |
| "eval_mcc": 0.5812582155687692, | |
| "eval_runtime": 3.1409, | |
| "eval_samples_per_second": 727.824, | |
| "eval_steps_per_second": 91.058, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "eval_accuracy": 0.8048993945121765, | |
| "eval_f1": 0.820883534136546, | |
| "eval_loss": 0.8462380170822144, | |
| "eval_mcc": 0.6077896979937155, | |
| "eval_runtime": 3.1222, | |
| "eval_samples_per_second": 732.17, | |
| "eval_steps_per_second": 91.601, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 1.5846994535519128e-05, | |
| "loss": 0.1045, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_accuracy": 0.8018372654914856, | |
| "eval_f1": 0.8210193599367839, | |
| "eval_loss": 0.9258682131767273, | |
| "eval_mcc": 0.6022610616438395, | |
| "eval_runtime": 3.1586, | |
| "eval_samples_per_second": 723.747, | |
| "eval_steps_per_second": 90.547, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 9.016393442622952e-06, | |
| "loss": 0.06, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "eval_accuracy": 0.8009623885154724, | |
| "eval_f1": 0.8171956609079951, | |
| "eval_loss": 1.0139225721359253, | |
| "eval_mcc": 0.5998412724259652, | |
| "eval_runtime": 3.1306, | |
| "eval_samples_per_second": 730.21, | |
| "eval_steps_per_second": 91.356, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 9.56, | |
| "learning_rate": 2.185792349726776e-06, | |
| "loss": 0.0406, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "eval_accuracy": 0.8070865869522095, | |
| "eval_f1": 0.8221056877773294, | |
| "eval_loss": 1.1088656187057495, | |
| "eval_mcc": 0.6121534397963299, | |
| "eval_runtime": 3.1448, | |
| "eval_samples_per_second": 726.919, | |
| "eval_steps_per_second": 90.944, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 9.84, | |
| "step": 3600, | |
| "total_flos": 2.842483414770432e+16, | |
| "train_loss": 0.24041106661160788, | |
| "train_runtime": 1450.9252, | |
| "train_samples_per_second": 302.683, | |
| "train_steps_per_second": 2.523 | |
| } | |
| ], | |
| "max_steps": 3660, | |
| "num_train_epochs": 10, | |
| "total_flos": 2.842483414770432e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |