{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 33,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 1e-05,
      "loss": 2.2461,
      "step": 1
    },
    {
      "epoch": 0.18,
      "learning_rate": 2e-05,
      "loss": 2.0254,
      "step": 2
    },
    {
      "epoch": 0.27,
      "learning_rate": 1.994869323391895e-05,
      "loss": 1.957,
      "step": 3
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.9795299412524948e-05,
      "loss": 1.8125,
      "step": 4
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.954139256400049e-05,
      "loss": 1.6719,
      "step": 5
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.918957811620231e-05,
      "loss": 1.6621,
      "step": 6
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.8743466161445823e-05,
      "loss": 1.8145,
      "step": 7
    },
    {
      "epoch": 0.73,
      "learning_rate": 1.8207634412072765e-05,
      "loss": 1.832,
      "step": 8
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.758758122692791e-05,
      "loss": 1.6309,
      "step": 9
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.688966919075687e-05,
      "loss": 1.916,
      "step": 10
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.612105982547663e-05,
      "loss": 1.168,
      "step": 11
    },
    {
      "epoch": 1.09,
      "learning_rate": 1.5289640103269626e-05,
      "loss": 0.6362,
      "step": 12
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.4403941515576344e-05,
      "loss": 0.7109,
      "step": 13
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.3473052528448203e-05,
      "loss": 0.6787,
      "step": 14
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.2506525322587207e-05,
      "loss": 0.5903,
      "step": 15
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.1514277775045768e-05,
      "loss": 0.4919,
      "step": 16
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.0506491688387128e-05,
      "loss": 0.4951,
      "step": 17
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.493508311612874e-06,
      "loss": 0.5684,
      "step": 18
    },
    {
      "epoch": 1.73,
      "learning_rate": 8.485722224954237e-06,
      "loss": 0.5947,
      "step": 19
    },
    {
      "epoch": 1.82,
      "learning_rate": 7.493474677412795e-06,
      "loss": 0.4961,
      "step": 20
    },
    {
      "epoch": 1.82,
      "eval_loss": 2.140625,
      "eval_runtime": 2.2551,
      "eval_samples_per_second": 0.443,
      "eval_steps_per_second": 0.443,
      "step": 20
    },
    {
      "epoch": 1.91,
      "learning_rate": 6.526947471551799e-06,
      "loss": 0.4053,
      "step": 21
    },
    {
      "epoch": 2.0,
      "learning_rate": 5.5960584844236565e-06,
      "loss": 0.3503,
      "step": 22
    },
    {
      "epoch": 2.09,
      "learning_rate": 4.710359896730379e-06,
      "loss": 0.2158,
      "step": 23
    },
    {
      "epoch": 2.18,
      "learning_rate": 3.878940174523371e-06,
      "loss": 0.1437,
      "step": 24
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.110330809243134e-06,
      "loss": 0.1854,
      "step": 25
    },
    {
      "epoch": 2.36,
      "learning_rate": 2.4124187730720916e-06,
      "loss": 0.1115,
      "step": 26
    },
    {
      "epoch": 2.45,
      "learning_rate": 1.7923655879272395e-06,
      "loss": 0.111,
      "step": 27
    },
    {
      "epoch": 2.55,
      "learning_rate": 1.2565338385541792e-06,
      "loss": 0.2671,
      "step": 28
    },
    {
      "epoch": 2.64,
      "learning_rate": 8.10421883797694e-07,
      "loss": 0.2568,
      "step": 29
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.5860743599951186e-07,
      "loss": 0.1287,
      "step": 30
    },
    {
      "epoch": 2.82,
      "learning_rate": 2.0470058747505516e-07,
      "loss": 0.1937,
      "step": 31
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.1306766081048456e-08,
      "loss": 0.1647,
      "step": 32
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 0.1019,
      "step": 33
    }
  ],
  "max_steps": 33,
  "num_train_epochs": 3,
  "total_flos": 652298158080.0,
  "trial_name": null,
  "trial_params": null
}