{ "best_metric": 0.8468692367302207, "best_model_checkpoint": "PhoBertPunc\\checkpoint-5746", "epoch": 13.0, "global_step": 5746, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 1.9002262443438914e-05, "loss": 0.3702, "step": 441 }, { "epoch": 1.0, "eval_accuracy": 0.6371049949031601, "eval_f1": 0.7381062355658198, "eval_loss": 0.285582959651947, "eval_roc_auc": 0.8207916729455642, "eval_runtime": 9.0929, "eval_samples_per_second": 107.886, "eval_steps_per_second": 5.499, "step": 442 }, { "epoch": 2.0, "learning_rate": 1.800452488687783e-05, "loss": 0.2734, "step": 882 }, { "epoch": 2.0, "eval_accuracy": 0.6605504587155964, "eval_f1": 0.7523277467411547, "eval_loss": 0.26946958899497986, "eval_roc_auc": 0.828197000778661, "eval_runtime": 7.7509, "eval_samples_per_second": 126.566, "eval_steps_per_second": 6.451, "step": 884 }, { "epoch": 2.99, "learning_rate": 1.7006787330316743e-05, "loss": 0.2267, "step": 1323 }, { "epoch": 3.0, "eval_accuracy": 0.6595310907237513, "eval_f1": 0.7564625850340136, "eval_loss": 0.27753132581710815, "eval_roc_auc": 0.8355058707497308, "eval_runtime": 7.6581, "eval_samples_per_second": 128.1, "eval_steps_per_second": 6.529, "step": 1326 }, { "epoch": 3.99, "learning_rate": 1.600904977375566e-05, "loss": 0.1876, "step": 1764 }, { "epoch": 4.0, "eval_accuracy": 0.6371049949031601, "eval_f1": 0.763459841129744, "eval_loss": 0.29114222526550293, "eval_roc_auc": 0.8451443088928317, "eval_runtime": 7.891, "eval_samples_per_second": 124.318, "eval_steps_per_second": 6.336, "step": 1768 }, { "epoch": 4.99, "learning_rate": 1.5011312217194571e-05, "loss": 0.1572, "step": 2205 }, { "epoch": 5.0, "eval_accuracy": 0.6371049949031601, "eval_f1": 0.7529827662395051, "eval_loss": 0.31292596459388733, "eval_roc_auc": 0.8381052176821842, "eval_runtime": 7.5397, "eval_samples_per_second": 130.111, "eval_steps_per_second": 6.632, "step": 2210 }, { "epoch": 5.99, "learning_rate": 1.4013574660633484e-05, "loss": 0.1312, "step": 2646 }, { "epoch": 6.0, "eval_accuracy": 0.6167176350662589, "eval_f1": 0.7478632478632479, "eval_loss": 0.3474615812301636, "eval_roc_auc": 0.8410431585366024, "eval_runtime": 8.8902, "eval_samples_per_second": 110.346, "eval_steps_per_second": 5.624, "step": 2652 }, { "epoch": 6.98, "learning_rate": 1.30158371040724e-05, "loss": 0.1048, "step": 3087 }, { "epoch": 7.0, "eval_accuracy": 0.6187563710499491, "eval_f1": 0.7460869565217391, "eval_loss": 0.3677310347557068, "eval_roc_auc": 0.8366267414318126, "eval_runtime": 7.6489, "eval_samples_per_second": 128.254, "eval_steps_per_second": 6.537, "step": 3094 }, { "epoch": 7.98, "learning_rate": 1.2018099547511313e-05, "loss": 0.0844, "step": 3528 }, { "epoch": 8.0, "eval_accuracy": 0.6034658511722731, "eval_f1": 0.7418244406196213, "eval_loss": 0.39172324538230896, "eval_roc_auc": 0.835729578342679, "eval_runtime": 7.4868, "eval_samples_per_second": 131.031, "eval_steps_per_second": 6.678, "step": 3536 }, { "epoch": 8.98, "learning_rate": 1.1020361990950229e-05, "loss": 0.0698, "step": 3969 }, { "epoch": 9.0, "eval_accuracy": 0.6106014271151886, "eval_f1": 0.7507481829841812, "eval_loss": 0.42587536573410034, "eval_roc_auc": 0.842892186936503, "eval_runtime": 7.5353, "eval_samples_per_second": 130.188, "eval_steps_per_second": 6.635, "step": 3978 }, { "epoch": 9.98, "learning_rate": 1.0022624434389141e-05, "loss": 0.054, "step": 4410 }, { "epoch": 10.0, "eval_accuracy": 0.5953109072375128, "eval_f1": 0.7450317124735729, "eval_loss": 0.4620385468006134, "eval_roc_auc": 0.8411574616862839, "eval_runtime": 7.8298, "eval_samples_per_second": 125.29, "eval_steps_per_second": 6.386, "step": 4420 }, { "epoch": 10.98, "learning_rate": 9.024886877828056e-06, "loss": 0.0451, "step": 4851 }, { "epoch": 11.0, "eval_accuracy": 0.6330275229357798, "eval_f1": 0.7534965034965035, "eval_loss": 0.45941799879074097, "eval_roc_auc": 0.8405079165428382, "eval_runtime": 7.6958, "eval_samples_per_second": 127.473, "eval_steps_per_second": 6.497, "step": 4862 }, { "epoch": 11.97, "learning_rate": 8.02714932126697e-06, "loss": 0.0366, "step": 5292 }, { "epoch": 12.0, "eval_accuracy": 0.5891946992864424, "eval_f1": 0.7466777408637875, "eval_loss": 0.5142910480499268, "eval_roc_auc": 0.8457477828688038, "eval_runtime": 7.657, "eval_samples_per_second": 128.118, "eval_steps_per_second": 6.53, "step": 5304 }, { "epoch": 12.97, "learning_rate": 7.029411764705882e-06, "loss": 0.0288, "step": 5733 }, { "epoch": 13.0, "eval_accuracy": 0.601427115188583, "eval_f1": 0.751892346509672, "eval_loss": 0.5171791315078735, "eval_roc_auc": 0.8468692367302207, "eval_runtime": 7.4917, "eval_samples_per_second": 130.944, "eval_steps_per_second": 6.674, "step": 5746 } ], "max_steps": 8840, "num_train_epochs": 20, "total_flos": 1.5096534982943232e+16, "trial_name": null, "trial_params": null }