{
  "best_metric": 0.5256543159484863,
  "best_model_checkpoint": "/tmp/model/checkpoint-39",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08,
      "learning_rate": 1.25e-05,
      "loss": 0.6934,
      "step": 1
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.5e-05,
      "loss": 0.6934,
      "step": 2
    },
    {
      "epoch": 0.23,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.6927,
      "step": 3
    },
    {
      "epoch": 0.31,
      "learning_rate": 5e-05,
      "loss": 0.6925,
      "step": 4
    },
    {
      "epoch": 0.38,
      "learning_rate": 5e-05,
      "loss": 0.6982,
      "step": 5
    },
    {
      "epoch": 0.46,
      "learning_rate": 4.8571428571428576e-05,
      "loss": 0.6873,
      "step": 6
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.714285714285714e-05,
      "loss": 0.6981,
      "step": 7
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.5714285714285716e-05,
      "loss": 0.6832,
      "step": 8
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.428571428571428e-05,
      "loss": 0.6876,
      "step": 9
    },
    {
      "epoch": 0.77,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.6754,
      "step": 10
    },
    {
      "epoch": 0.85,
      "learning_rate": 4.1428571428571437e-05,
      "loss": 0.665,
      "step": 11
    },
    {
      "epoch": 0.92,
      "learning_rate": 4e-05,
      "loss": 0.6749,
      "step": 12
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.857142857142858e-05,
      "loss": 0.6663,
      "step": 13
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.56,
      "eval_auc": 0.922077922077922,
      "eval_f1": 0.717948717948718,
      "eval_loss": 0.6754491925239563,
      "eval_precision": 0.56,
      "eval_recall": 1.0,
      "eval_runtime": 1.3736,
      "eval_samples_per_second": 18.2,
      "eval_steps_per_second": 1.456,
      "step": 13
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.7142857142857143e-05,
      "loss": 0.6102,
      "step": 14
    },
    {
      "epoch": 1.15,
      "learning_rate": 3.571428571428572e-05,
      "loss": 0.6307,
      "step": 15
    },
    {
      "epoch": 1.23,
      "learning_rate": 3.428571428571429e-05,
      "loss": 0.6337,
      "step": 16
    },
    {
      "epoch": 1.31,
      "learning_rate": 3.285714285714286e-05,
      "loss": 0.6769,
      "step": 17
    },
    {
      "epoch": 1.38,
      "learning_rate": 3.142857142857143e-05,
      "loss": 0.6445,
      "step": 18
    },
    {
      "epoch": 1.46,
      "learning_rate": 3e-05,
      "loss": 0.7123,
      "step": 19
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.6527,
      "step": 20
    },
    {
      "epoch": 1.62,
      "learning_rate": 2.714285714285714e-05,
      "loss": 0.69,
      "step": 21
    },
    {
      "epoch": 1.69,
      "learning_rate": 2.5714285714285714e-05,
      "loss": 0.6184,
      "step": 22
    },
    {
      "epoch": 1.77,
      "learning_rate": 2.4285714285714288e-05,
      "loss": 0.6183,
      "step": 23
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.2857142857142858e-05,
      "loss": 0.6548,
      "step": 24
    },
    {
      "epoch": 1.92,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.6148,
      "step": 25
    },
    {
      "epoch": 2.0,
      "learning_rate": 2e-05,
      "loss": 0.5487,
      "step": 26
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.84,
      "eval_auc": 0.974025974025974,
      "eval_f1": 0.8333333333333333,
      "eval_loss": 0.5991113185882568,
      "eval_precision": 1.0,
      "eval_recall": 0.7142857142857143,
      "eval_runtime": 1.2936,
      "eval_samples_per_second": 19.326,
      "eval_steps_per_second": 1.546,
      "step": 26
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.8571428571428572e-05,
      "loss": 0.6147,
      "step": 27
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.7142857142857145e-05,
      "loss": 0.6138,
      "step": 28
    },
    {
      "epoch": 2.23,
      "learning_rate": 1.5714285714285715e-05,
      "loss": 0.6359,
      "step": 29
    },
    {
      "epoch": 2.31,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.574,
      "step": 30
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.2857142857142857e-05,
      "loss": 0.537,
      "step": 31
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.5474,
      "step": 32
    },
    {
      "epoch": 2.54,
      "learning_rate": 1e-05,
      "loss": 0.6105,
      "step": 33
    },
    {
      "epoch": 2.62,
      "learning_rate": 8.571428571428573e-06,
      "loss": 0.5906,
      "step": 34
    },
    {
      "epoch": 2.69,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.5601,
      "step": 35
    },
    {
      "epoch": 2.77,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.4603,
      "step": 36
    },
    {
      "epoch": 2.85,
      "learning_rate": 4.285714285714286e-06,
      "loss": 0.5096,
      "step": 37
    },
    {
      "epoch": 2.92,
      "learning_rate": 2.8571428571428573e-06,
      "loss": 0.5837,
      "step": 38
    },
    {
      "epoch": 3.0,
      "learning_rate": 1.4285714285714286e-06,
      "loss": 0.4309,
      "step": 39
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.92,
      "eval_auc": 0.9805194805194806,
      "eval_f1": 0.9285714285714286,
      "eval_loss": 0.5256543159484863,
      "eval_precision": 0.9285714285714286,
      "eval_recall": 0.9285714285714286,
      "eval_runtime": 1.3109,
      "eval_samples_per_second": 19.071,
      "eval_steps_per_second": 1.526,
      "step": 39
    }
  ],
  "logging_steps": 1,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 8.135845141977907e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}