{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 16599,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 4.849388517380565e-05,
      "loss": 2.4052,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.6987770347611306e-05,
      "loss": 1.6715,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.5481655521416954e-05,
      "loss": 1.5473,
      "step": 1500
    },
    {
      "epoch": 0.36,
      "learning_rate": 4.397554069522261e-05,
      "loss": 1.4151,
      "step": 2000
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.246942586902826e-05,
      "loss": 1.3605,
      "step": 2500
    },
    {
      "epoch": 0.54,
      "learning_rate": 4.0963311042833905e-05,
      "loss": 1.3104,
      "step": 3000
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.945719621663956e-05,
      "loss": 1.2118,
      "step": 3500
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.795108139044521e-05,
      "loss": 1.2702,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "learning_rate": 3.644496656425086e-05,
      "loss": 1.2225,
      "step": 4500
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.493885173805651e-05,
      "loss": 1.1811,
      "step": 5000
    },
    {
      "epoch": 0.99,
      "learning_rate": 3.343273691186216e-05,
      "loss": 1.1898,
      "step": 5500
    },
    {
      "epoch": 1.08,
      "learning_rate": 3.1926622085667815e-05,
      "loss": 0.929,
      "step": 6000
    },
    {
      "epoch": 1.17,
      "learning_rate": 3.0420507259473463e-05,
      "loss": 0.8824,
      "step": 6500
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.8914392433279115e-05,
      "loss": 0.8851,
      "step": 7000
    },
    {
      "epoch": 1.36,
      "learning_rate": 2.7408277607084763e-05,
      "loss": 0.9067,
      "step": 7500
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.5902162780890415e-05,
      "loss": 0.9111,
      "step": 8000
    },
    {
      "epoch": 1.54,
      "learning_rate": 2.4396047954696066e-05,
      "loss": 0.8716,
      "step": 8500
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.2889933128501718e-05,
      "loss": 0.8914,
      "step": 9000
    },
    {
      "epoch": 1.72,
      "learning_rate": 2.138381830230737e-05,
      "loss": 0.8454,
      "step": 9500
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.9877703476113018e-05,
      "loss": 0.8956,
      "step": 10000
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.837158864991867e-05,
      "loss": 0.8595,
      "step": 10500
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.686547382372432e-05,
      "loss": 0.8699,
      "step": 11000
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.5359358997529973e-05,
      "loss": 0.6436,
      "step": 11500
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.3853244171335622e-05,
      "loss": 0.6209,
      "step": 12000
    },
    {
      "epoch": 2.26,
      "learning_rate": 1.2347129345141274e-05,
      "loss": 0.6164,
      "step": 12500
    },
    {
      "epoch": 2.35,
      "learning_rate": 1.0841014518946926e-05,
      "loss": 0.6244,
      "step": 13000
    },
    {
      "epoch": 2.44,
      "learning_rate": 9.334899692752576e-06,
      "loss": 0.6159,
      "step": 13500
    },
    {
      "epoch": 2.53,
      "learning_rate": 7.828784866558227e-06,
      "loss": 0.6309,
      "step": 14000
    },
    {
      "epoch": 2.62,
      "learning_rate": 6.322670040363877e-06,
      "loss": 0.617,
      "step": 14500
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.816555214169529e-06,
      "loss": 0.615,
      "step": 15000
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.310440387975179e-06,
      "loss": 0.5898,
      "step": 15500
    },
    {
      "epoch": 2.89,
      "learning_rate": 1.8043255617808302e-06,
      "loss": 0.6219,
      "step": 16000
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.982107355864811e-07,
      "loss": 0.6043,
      "step": 16500
    },
    {
      "epoch": 3.0,
      "step": 16599,
      "total_flos": 1.5432246862350336e+16,
      "train_loss": 0.9777237792009043,
      "train_runtime": 2351.827,
      "train_samples_per_second": 112.922,
      "train_steps_per_second": 7.058
    }
  ],
  "max_steps": 16599,
  "num_train_epochs": 3,
  "total_flos": 1.5432246862350336e+16,
  "trial_name": null,
  "trial_params": null
}