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": 1.6673333333333335e-05,
      "loss": 0.4103,
      "step": 500
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3346666666666667e-05,
      "loss": 0.3177,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.886,
      "eval_classification_report": "              precision    recall  f1-score   support\n0              0.897959  0.958416  0.927203  3030.000\n1              0.835509  0.659794  0.737327   970.000\naccuracy       0.886000  0.886000  0.886000     0.886\nmacro avg      0.866734  0.809105  0.832265  4000.000\nweighted avg   0.882815  0.886000  0.881158  4000.000",
      "eval_confusion_matrix": "[[2904  126]\n [ 330  640]]",
      "eval_confusion_matrix_norm": "[[0.95841584 0.04158416]\n [0.34020619 0.65979381]]",
      "eval_f1": 0.7373271889400921,
      "eval_f1_macro": 0.8322651270370958,
      "eval_f1_weighted": 0.8811581651570527,
      "eval_loss": 0.28936144709587097,
      "eval_runtime": 9.6533,
      "eval_samples_per_second": 414.366,
      "eval_steps_per_second": 25.898,
      "step": 1000
    },
    {
      "epoch": 1.5,
      "learning_rate": 1.0013333333333335e-05,
      "loss": 0.2395,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.680000000000001e-06,
      "loss": 0.2232,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.88325,
      "eval_classification_report": "              precision    recall  f1-score     support\n0              0.921407  0.924752  0.923077  3030.00000\n1              0.762252  0.753608  0.757906   970.00000\naccuracy       0.883250  0.883250  0.883250     0.88325\nmacro avg      0.841830  0.839180  0.840491  4000.00000\nweighted avg   0.882812  0.883250  0.883023  4000.00000",
      "eval_confusion_matrix": "[[2802  228]\n [ 239  731]]",
      "eval_confusion_matrix_norm": "[[0.92475248 0.07524752]\n [0.24639175 0.75360825]]",
      "eval_f1": 0.7579056505961639,
      "eval_f1_macro": 0.8404912868365435,
      "eval_f1_weighted": 0.883022889500339,
      "eval_loss": 0.3370012938976288,
      "eval_runtime": 9.6971,
      "eval_samples_per_second": 412.493,
      "eval_steps_per_second": 25.781,
      "step": 2000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.346666666666667e-06,
      "loss": 0.1563,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6666666666666667e-08,
      "loss": 0.1534,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.887,
      "eval_classification_report": "              precision    recall  f1-score   support\n0              0.924013  0.927063  0.925535  3030.000\n1              0.769792  0.761856  0.765803   970.000\naccuracy       0.887000  0.887000  0.887000     0.887\nmacro avg      0.846902  0.844459  0.845669  4000.000\nweighted avg   0.886614  0.887000  0.886800  4000.000",
      "eval_confusion_matrix": "[[2809  221]\n [ 231  739]]",
      "eval_confusion_matrix_norm": "[[0.92706271 0.07293729]\n [0.23814433 0.76185567]]",
      "eval_f1": 0.7658031088082901,
      "eval_f1_macro": 0.8456692644535684,
      "eval_f1_weighted": 0.8868003346108867,
      "eval_loss": 0.42191165685653687,
      "eval_runtime": 9.677,
      "eval_samples_per_second": 413.351,
      "eval_steps_per_second": 25.834,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "step": 3000,
      "total_flos": 1.262933065728e+16,
      "train_loss": 0.25008199564615885,
      "train_runtime": 436.7193,
      "train_samples_per_second": 109.91,
      "train_steps_per_second": 6.869
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 3,
  "total_flos": 1.262933065728e+16,
  "trial_name": null,
  "trial_params": null
}