checkpoint-2500/trainer_state.json

{
  "best_metric": 0.32178911566734314,
  "best_model_checkpoint": "/tmp/model/checkpoint-2500",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "learning_rate": 8.266666666666667e-06,
      "loss": 0.6965,
      "step": 125
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.6533333333333333e-05,
      "loss": 0.5176,
      "step": 250
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.486666666666667e-05,
      "loss": 0.4482,
      "step": 375
    },
    {
      "epoch": 0.2,
      "learning_rate": 3.313333333333333e-05,
      "loss": 0.4278,
      "step": 500
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.146666666666667e-05,
      "loss": 0.4267,
      "step": 625
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.973333333333334e-05,
      "loss": 0.4106,
      "step": 750
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.910370370370371e-05,
      "loss": 0.4548,
      "step": 875
    },
    {
      "epoch": 0.4,
      "learning_rate": 4.817777777777778e-05,
      "loss": 0.3838,
      "step": 1000
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.725185185185185e-05,
      "loss": 0.3781,
      "step": 1125
    },
    {
      "epoch": 0.5,
      "learning_rate": 4.6325925925925926e-05,
      "loss": 0.4206,
      "step": 1250
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.5400000000000006e-05,
      "loss": 0.3933,
      "step": 1375
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.447407407407407e-05,
      "loss": 0.4357,
      "step": 1500
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.354814814814815e-05,
      "loss": 0.3704,
      "step": 1625
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.2622222222222224e-05,
      "loss": 0.3859,
      "step": 1750
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.16962962962963e-05,
      "loss": 0.3682,
      "step": 1875
    },
    {
      "epoch": 0.8,
      "learning_rate": 4.077037037037037e-05,
      "loss": 0.3852,
      "step": 2000
    },
    {
      "epoch": 0.85,
      "learning_rate": 3.984444444444444e-05,
      "loss": 0.3641,
      "step": 2125
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.891851851851852e-05,
      "loss": 0.3814,
      "step": 2250
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.7992592592592596e-05,
      "loss": 0.3471,
      "step": 2375
    },
    {
      "epoch": 1.0,
      "learning_rate": 3.706666666666667e-05,
      "loss": 0.3565,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8518,
      "eval_auc": 0.94403024,
      "eval_f1": 0.8431746031746032,
      "eval_loss": 0.32178911566734314,
      "eval_precision": 0.8952808988764045,
      "eval_recall": 0.7968,
      "eval_runtime": 10.5615,
      "eval_samples_per_second": 473.418,
      "eval_steps_per_second": 29.636,
      "step": 2500
    }
  ],
  "logging_steps": 125,
  "max_steps": 7500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1315555276800000.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}