{
  "best_metric": 0.03554883599281311,
  "best_model_checkpoint": "MD_TL_best_model/checkpoint-4089",
  "epoch": 3.0,
  "global_step": 4089,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37,
      "learning_rate": 1.8777207141110298e-05,
      "loss": 0.1048,
      "step": 500
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.7554414282220594e-05,
      "loss": 0.0502,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_balanced accuracy": 0.7151924088243289,
      "eval_f1": 0.7168079195427047,
      "eval_loss": 0.04138244688510895,
      "eval_precision": 0.7342695204092529,
      "eval_recall": 0.7151924088243289,
      "eval_runtime": 9.9389,
      "eval_samples_per_second": 548.553,
      "eval_steps_per_second": 34.31,
      "step": 1363
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.633162142333089e-05,
      "loss": 0.0401,
      "step": 1500
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.5108828564441186e-05,
      "loss": 0.0326,
      "step": 2000
    },
    {
      "epoch": 1.83,
      "learning_rate": 1.3886035705551482e-05,
      "loss": 0.0311,
      "step": 2500
    },
    {
      "epoch": 2.0,
      "eval_balanced accuracy": 0.7398895952902634,
      "eval_f1": 0.7495616067630219,
      "eval_loss": 0.03639577701687813,
      "eval_precision": 0.7655480981233173,
      "eval_recall": 0.7398895952902634,
      "eval_runtime": 4.3809,
      "eval_samples_per_second": 1244.494,
      "eval_steps_per_second": 77.838,
      "step": 2726
    },
    {
      "epoch": 2.2,
      "learning_rate": 1.2663242846661777e-05,
      "loss": 0.025,
      "step": 3000
    },
    {
      "epoch": 2.57,
      "learning_rate": 1.1440449987772073e-05,
      "loss": 0.0222,
      "step": 3500
    },
    {
      "epoch": 2.93,
      "learning_rate": 1.0217657128882368e-05,
      "loss": 0.0222,
      "step": 4000
    },
    {
      "epoch": 3.0,
      "eval_balanced accuracy": 0.7662314481801649,
      "eval_f1": 0.7739129932274338,
      "eval_loss": 0.03554883599281311,
      "eval_precision": 0.7868185694908753,
      "eval_recall": 0.7662314481801649,
      "eval_runtime": 4.3603,
      "eval_samples_per_second": 1250.36,
      "eval_steps_per_second": 78.205,
      "step": 4089
    }
  ],
  "max_steps": 8178,
  "num_train_epochs": 6,
  "total_flos": 355494913244352.0,
  "trial_name": null,
  "trial_params": null
}