trainer_state.json

{
  "best_metric": 2.4025585651397705,
  "best_model_checkpoint": "output/tool/checkpoint-55",
  "epoch": 5.0,
  "global_step": 55,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.45,
      "learning_rate": 7.836279790554737e-05,
      "loss": 3.2081,
      "step": 5
    },
    {
      "epoch": 0.91,
      "learning_rate": 2.778782010045488e-06,
      "loss": 3.1621,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.746370792388916,
      "eval_runtime": 0.6285,
      "eval_samples_per_second": 22.274,
      "eval_steps_per_second": 3.182,
      "step": 11
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.0102530108070535e-05,
      "loss": 2.9659,
      "step": 15
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.000126309992352219,
      "loss": 2.8367,
      "step": 20
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.761970043182373,
      "eval_runtime": 0.6295,
      "eval_samples_per_second": 22.239,
      "eval_steps_per_second": 3.177,
      "step": 22
    },
    {
      "epoch": 2.27,
      "learning_rate": 0.00011352344634864655,
      "loss": 2.6931,
      "step": 25
    },
    {
      "epoch": 2.73,
      "learning_rate": 2.3676553651353525e-05,
      "loss": 2.6297,
      "step": 30
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.716337203979492,
      "eval_runtime": 0.6316,
      "eval_samples_per_second": 22.166,
      "eval_steps_per_second": 3.167,
      "step": 33
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.0890007647780947e-05,
      "loss": 2.7361,
      "step": 35
    },
    {
      "epoch": 3.64,
      "learning_rate": 9.709746989192932e-05,
      "loss": 2.4185,
      "step": 40
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.619758367538452,
      "eval_runtime": 0.5572,
      "eval_samples_per_second": 23.33,
      "eval_steps_per_second": 3.589,
      "step": 44
    },
    {
      "epoch": 4.09,
      "learning_rate": 0.00013442121798995453,
      "loss": 2.4054,
      "step": 45
    },
    {
      "epoch": 4.55,
      "learning_rate": 5.883720209445262e-05,
      "loss": 2.2843,
      "step": 50
    },
    {
      "epoch": 5.0,
      "learning_rate": 0.0,
      "loss": 2.4935,
      "step": 55
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.4025585651397705,
      "eval_runtime": 0.1869,
      "eval_samples_per_second": 80.273,
      "eval_steps_per_second": 10.703,
      "step": 55
    }
  ],
  "max_steps": 55,
  "num_train_epochs": 5,
  "total_flos": 57222955008000.0,
  "trial_name": null,
  "trial_params": null
}