{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 9400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.32,
      "learning_rate": 4.734042553191489e-05,
      "loss": 1.3936,
      "step": 500
    },
    {
      "epoch": 10.64,
      "learning_rate": 4.468085106382979e-05,
      "loss": 0.1574,
      "step": 1000
    },
    {
      "epoch": 15.96,
      "learning_rate": 4.2021276595744684e-05,
      "loss": 0.0664,
      "step": 1500
    },
    {
      "epoch": 21.28,
      "learning_rate": 3.936170212765958e-05,
      "loss": 0.0374,
      "step": 2000
    },
    {
      "epoch": 26.6,
      "learning_rate": 3.670212765957447e-05,
      "loss": 0.0301,
      "step": 2500
    },
    {
      "epoch": 31.91,
      "learning_rate": 3.4042553191489365e-05,
      "loss": 0.0274,
      "step": 3000
    },
    {
      "epoch": 37.23,
      "learning_rate": 3.1382978723404254e-05,
      "loss": 0.0274,
      "step": 3500
    },
    {
      "epoch": 42.55,
      "learning_rate": 2.8723404255319154e-05,
      "loss": 0.0361,
      "step": 4000
    },
    {
      "epoch": 47.87,
      "learning_rate": 2.6063829787234046e-05,
      "loss": 0.0581,
      "step": 4500
    },
    {
      "epoch": 53.19,
      "learning_rate": 2.340425531914894e-05,
      "loss": 0.0763,
      "step": 5000
    },
    {
      "epoch": 58.51,
      "learning_rate": 2.074468085106383e-05,
      "loss": 0.1021,
      "step": 5500
    },
    {
      "epoch": 63.83,
      "learning_rate": 1.8085106382978724e-05,
      "loss": 0.1803,
      "step": 6000
    },
    {
      "epoch": 69.15,
      "learning_rate": 1.5425531914893617e-05,
      "loss": 0.2528,
      "step": 6500
    },
    {
      "epoch": 74.47,
      "learning_rate": 1.2765957446808511e-05,
      "loss": 0.264,
      "step": 7000
    },
    {
      "epoch": 79.79,
      "learning_rate": 1.0106382978723404e-05,
      "loss": 0.3108,
      "step": 7500
    },
    {
      "epoch": 85.11,
      "learning_rate": 7.446808510638298e-06,
      "loss": 0.1397,
      "step": 8000
    },
    {
      "epoch": 90.43,
      "learning_rate": 4.787234042553191e-06,
      "loss": 0.1356,
      "step": 8500
    },
    {
      "epoch": 95.74,
      "learning_rate": 2.1276595744680853e-06,
      "loss": 0.0587,
      "step": 9000
    },
    {
      "epoch": 100.0,
      "step": 9400,
      "total_flos": 2.93969475072e+16,
      "train_loss": 0.18034995332677314,
      "train_runtime": 1545.8589,
      "train_samples_per_second": 194.067,
      "train_steps_per_second": 6.081
    }
  ],
  "logging_steps": 500,
  "max_steps": 9400,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 2.93969475072e+16,
  "trial_name": null,
  "trial_params": null
}