training_checkpoints/checkpoint-1344/trainer_state.json

{
  "best_metric": 0.6299493260025852,
  "best_model_checkpoint": "case-analysis-bert-base-uncased/checkpoint-896",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1344,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7861915367483296,
      "eval_f1_macro": 0.5606209513919795,
      "eval_f1_micro": 0.7861915367483298,
      "eval_f1_weighted": 0.7596863457967054,
      "eval_loss": 0.7283128499984741,
      "eval_macro_fpr": 0.1142150114777894,
      "eval_macro_sensitivity": 0.5571958513134984,
      "eval_macro_specificity": 0.8974437739369907,
      "eval_precision": 0.7487100743874461,
      "eval_precision_macro": 0.584767668299669,
      "eval_recall": 0.7861915367483296,
      "eval_recall_macro": 0.5571958513134984,
      "eval_runtime": 9.7006,
      "eval_samples_per_second": 46.286,
      "eval_steps_per_second": 5.876,
      "eval_weighted_fpr": 0.08311688311688312,
      "eval_weighted_sensitivity": 0.7861915367483296,
      "eval_weighted_specificity": 0.8035835589996332,
      "step": 224
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.799554565701559,
      "eval_f1_macro": 0.5913567856292314,
      "eval_f1_micro": 0.799554565701559,
      "eval_f1_weighted": 0.7794312570974847,
      "eval_loss": 0.816001296043396,
      "eval_macro_fpr": 0.09972743460349152,
      "eval_macro_sensitivity": 0.6065148418089594,
      "eval_macro_specificity": 0.910318444398066,
      "eval_precision": 0.7603126937076594,
      "eval_precision_macro": 0.5769576335614072,
      "eval_recall": 0.799554565701559,
      "eval_recall_macro": 0.6065148418089594,
      "eval_runtime": 39.2963,
      "eval_samples_per_second": 11.426,
      "eval_steps_per_second": 1.451,
      "eval_weighted_fpr": 0.07712082262210797,
      "eval_weighted_sensitivity": 0.799554565701559,
      "eval_weighted_specificity": 0.8417192118907049,
      "step": 448
    },
    {
      "epoch": 2.232142857142857,
      "grad_norm": 10.089143753051758,
      "learning_rate": 4.6302083333333335e-05,
      "loss": 0.6512,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7906458797327395,
      "eval_f1_macro": 0.5840451851613266,
      "eval_f1_micro": 0.7906458797327395,
      "eval_f1_weighted": 0.771952482046502,
      "eval_loss": 0.8587548136711121,
      "eval_macro_fpr": 0.10045553338392213,
      "eval_macro_sensitivity": 0.5989408048231577,
      "eval_macro_specificity": 0.9104668136432382,
      "eval_precision": 0.7597703984690992,
      "eval_precision_macro": 0.5770004879708264,
      "eval_recall": 0.7906458797327395,
      "eval_recall_macro": 0.5989408048231577,
      "eval_runtime": 36.0068,
      "eval_samples_per_second": 12.47,
      "eval_steps_per_second": 1.583,
      "eval_weighted_fpr": 0.0811044003451251,
      "eval_weighted_sensitivity": 0.7906458797327395,
      "eval_weighted_specificity": 0.8512213748402133,
      "step": 672
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7817371937639198,
      "eval_f1_macro": 0.6299493260025852,
      "eval_f1_micro": 0.7817371937639198,
      "eval_f1_weighted": 0.7804892003577936,
      "eval_loss": 1.0820659399032593,
      "eval_macro_fpr": 0.09955325556886446,
      "eval_macro_sensitivity": 0.6429130620307091,
      "eval_macro_specificity": 0.9124114631310538,
      "eval_precision": 0.7819490155060428,
      "eval_precision_macro": 0.6213857962213225,
      "eval_recall": 0.7817371937639198,
      "eval_recall_macro": 0.6429130620307091,
      "eval_runtime": 32.7545,
      "eval_samples_per_second": 13.708,
      "eval_steps_per_second": 1.74,
      "eval_weighted_fpr": 0.08514335360556038,
      "eval_weighted_sensitivity": 0.7817371937639198,
      "eval_weighted_specificity": 0.8679086587602954,
      "step": 896
    },
    {
      "epoch": 4.464285714285714,
      "grad_norm": 0.3640059232711792,
      "learning_rate": 4.258184523809524e-05,
      "loss": 0.3466,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8084632516703786,
      "eval_f1_macro": 0.619470115868435,
      "eval_f1_micro": 0.8084632516703786,
      "eval_f1_weighted": 0.7927603955642376,
      "eval_loss": 1.0612273216247559,
      "eval_macro_fpr": 0.09480480467421379,
      "eval_macro_sensitivity": 0.6262706556824204,
      "eval_macro_specificity": 0.9138770264258179,
      "eval_precision": 0.7999415514782998,
      "eval_precision_macro": 0.7128659611992946,
      "eval_recall": 0.8084632516703786,
      "eval_recall_macro": 0.6262706556824204,
      "eval_runtime": 34.7281,
      "eval_samples_per_second": 12.929,
      "eval_steps_per_second": 1.641,
      "eval_weighted_fpr": 0.07319148936170213,
      "eval_weighted_sensitivity": 0.8084632516703786,
      "eval_weighted_specificity": 0.847044854032893,
      "step": 1120
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7928730512249443,
      "eval_f1_macro": 0.6273476992208005,
      "eval_f1_micro": 0.7928730512249443,
      "eval_f1_weighted": 0.7896519709076985,
      "eval_loss": 1.2559400796890259,
      "eval_macro_fpr": 0.09508832825281638,
      "eval_macro_sensitivity": 0.6361528640940406,
      "eval_macro_specificity": 0.9161350599199577,
      "eval_precision": 0.7876580739535682,
      "eval_precision_macro": 0.6205792331164909,
      "eval_recall": 0.7928730512249443,
      "eval_recall_macro": 0.6361528640940406,
      "eval_runtime": 38.0322,
      "eval_samples_per_second": 11.806,
      "eval_steps_per_second": 1.499,
      "eval_weighted_fpr": 0.08010335917312661,
      "eval_weighted_sensitivity": 0.7928730512249443,
      "eval_weighted_specificity": 0.8716671884548861,
      "step": 1344
    }
  ],
  "logging_steps": 500,
  "max_steps": 6720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 2829020867592192.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}