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": 1.6666666666666667e-05,
      "loss": 0.4103,
      "step": 500
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.3362,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.87225,
      "eval_classification_report": "              precision    recall  f1-score     support\n0              0.899714  0.935644  0.917327  3030.00000\n1              0.770318  0.674227  0.719076   970.00000\naccuracy       0.872250  0.872250  0.872250     0.87225\nmacro avg      0.835016  0.804935  0.818202  4000.00000\nweighted avg   0.868336  0.872250  0.869251  4000.00000",
      "eval_confusion_matrix": "[[2835  195]\n [ 316  654]]",
      "eval_confusion_matrix_norm": "[[0.93564356 0.06435644]\n [0.3257732  0.6742268 ]]",
      "eval_f1": 0.7190764156129743,
      "eval_f1_macro": 0.8182018544656038,
      "eval_f1_weighted": 0.8692514554747078,
      "eval_loss": 0.3033996522426605,
      "eval_runtime": 16.7014,
      "eval_samples_per_second": 239.501,
      "eval_steps_per_second": 14.969,
      "step": 1000
    },
    {
      "epoch": 1.5,
      "learning_rate": 1e-05,
      "loss": 0.2538,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.2352,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.87325,
      "eval_classification_report": "              precision    recall  f1-score     support\n0              0.915104  0.917822  0.916461  3030.00000\n1              0.740895  0.734021  0.737442   970.00000\naccuracy       0.873250  0.873250  0.873250     0.87325\nmacro avg      0.827999  0.825921  0.826951  4000.00000\nweighted avg   0.872858  0.873250  0.873049  4000.00000",
      "eval_confusion_matrix": "[[2781  249]\n [ 258  712]]",
      "eval_confusion_matrix_norm": "[[0.91782178 0.08217822]\n [0.26597938 0.73402062]]",
      "eval_f1": 0.737441740031072,
      "eval_f1_macro": 0.826951220979451,
      "eval_f1_weighted": 0.8730486036678663,
      "eval_loss": 0.37301740050315857,
      "eval_runtime": 16.7066,
      "eval_samples_per_second": 239.426,
      "eval_steps_per_second": 14.964,
      "step": 2000
    },
    {
      "epoch": 2.5,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 0.1625,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 0.1566,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8775,
      "eval_classification_report": "              precision    recall  f1-score    support\n0              0.906791  0.934323  0.920351  3030.0000\n1              0.773349  0.700000  0.734848   970.0000\naccuracy       0.877500  0.877500  0.877500     0.8775\nmacro avg      0.840070  0.817162  0.827600  4000.0000\nweighted avg   0.874431  0.877500  0.875367  4000.0000",
      "eval_confusion_matrix": "[[2831  199]\n [ 291  679]]",
      "eval_confusion_matrix_norm": "[[0.93432343 0.06567657]\n [0.3        0.7       ]]",
      "eval_f1": 0.7348484848484848,
      "eval_f1_macro": 0.8275997950900422,
      "eval_f1_weighted": 0.8753667198644444,
      "eval_loss": 0.4632544219493866,
      "eval_runtime": 16.6967,
      "eval_samples_per_second": 239.568,
      "eval_steps_per_second": 14.973,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "step": 3000,
      "total_flos": 1.262933065728e+16,
      "train_loss": 0.2591003138224284,
      "train_runtime": 651.1299,
      "train_samples_per_second": 73.718,
      "train_steps_per_second": 4.607
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 3,
  "total_flos": 1.262933065728e+16,
  "trial_name": null,
  "trial_params": null
}