trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "learning_rate": 1.6686666666666667e-05,
      "loss": 0.4418,
      "step": 500
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3353333333333333e-05,
      "loss": 0.3253,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.878,
      "eval_classification_report": "              precision    recall  f1-score   support\n0              0.901961  0.941254  0.921189  3030.000\n1              0.787589  0.680412  0.730088   970.000\naccuracy       0.878000  0.878000  0.878000     0.878\nmacro avg      0.844775  0.810833  0.825639  4000.000\nweighted avg   0.874226  0.878000  0.874847  4000.000",
      "eval_confusion_matrix": "[[2852  178]\n [ 310  660]]",
      "eval_confusion_matrix_norm": "[[0.94125413 0.05874587]\n [0.31958763 0.68041237]]",
      "eval_f1": 0.7300884955752213,
      "eval_f1_macro": 0.8256385630330887,
      "eval_f1_weighted": 0.8748468477738904,
      "eval_loss": 0.3010723888874054,
      "eval_runtime": 9.6596,
      "eval_samples_per_second": 414.097,
      "eval_steps_per_second": 25.881,
      "step": 1000
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.002e-05,
      "loss": 0.2684,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.6866666666666665e-06,
      "loss": 0.2439,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8865,
      "eval_classification_report": "              precision    recall  f1-score    support\n0              0.914414  0.937954  0.926035  3030.0000\n1              0.789238  0.725773  0.756176   970.0000\naccuracy       0.886500  0.886500  0.886500     0.8865\nmacro avg      0.851826  0.831863  0.841105  4000.0000\nweighted avg   0.884059  0.886500  0.884844  4000.0000",
      "eval_confusion_matrix": "[[2842  188]\n [ 266  704]]",
      "eval_confusion_matrix_norm": "[[0.9379538 0.0620462]\n [0.2742268 0.7257732]]",
      "eval_f1": 0.7561761546723953,
      "eval_f1_macro": 0.84110534680508,
      "eval_f1_weighted": 0.8848438807534127,
      "eval_loss": 0.3121766746044159,
      "eval_runtime": 9.6642,
      "eval_samples_per_second": 413.897,
      "eval_steps_per_second": 25.869,
      "step": 2000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3600000000000004e-06,
      "loss": 0.2023,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6666666666666667e-08,
      "loss": 0.1962,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.88075,
      "eval_classification_report": "              precision    recall  f1-score     support\n0              0.920593  0.922112  0.921352  3030.00000\n1              0.755440  0.751546  0.753488   970.00000\naccuracy       0.880750  0.880750  0.880750     0.88075\nmacro avg      0.838017  0.836829  0.837420  4000.00000\nweighted avg   0.880544  0.880750  0.880645  4000.00000",
      "eval_confusion_matrix": "[[2794  236]\n [ 241  729]]",
      "eval_confusion_matrix_norm": "[[0.92211221 0.07788779]\n [0.24845361 0.75154639]]",
      "eval_f1": 0.7534883720930233,
      "eval_f1_macro": 0.8374201959393394,
      "eval_f1_weighted": 0.8806450852201921,
      "eval_loss": 0.3457934856414795,
      "eval_runtime": 9.6651,
      "eval_samples_per_second": 413.86,
      "eval_steps_per_second": 25.866,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "step": 3000,
      "total_flos": 1.262933065728e+16,
      "train_loss": 0.27966182454427085,
      "train_runtime": 430.7509,
      "train_samples_per_second": 111.433,
      "train_steps_per_second": 6.965
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 3,
  "total_flos": 1.262933065728e+16,
  "trial_name": null,
  "trial_params": null
}