{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "global_step": 9656,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1,
      "learning_rate": 4.7410936205468105e-05,
      "loss": 0.0291,
      "step": 500
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.482187241093621e-05,
      "loss": 0.0114,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.2232808616404316e-05,
      "loss": 0.0126,
      "step": 1500
    },
    {
      "epoch": 0.41,
      "learning_rate": 3.964374482187241e-05,
      "loss": 0.0514,
      "step": 2000
    },
    {
      "epoch": 0.52,
      "learning_rate": 3.705468102734051e-05,
      "loss": 0.0156,
      "step": 2500
    },
    {
      "epoch": 0.62,
      "learning_rate": 3.4465617232808616e-05,
      "loss": 0.0137,
      "step": 3000
    },
    {
      "epoch": 0.72,
      "learning_rate": 3.187655343827672e-05,
      "loss": 0.0089,
      "step": 3500
    },
    {
      "epoch": 0.83,
      "learning_rate": 2.9287489643744827e-05,
      "loss": 0.0093,
      "step": 4000
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.6698425849212926e-05,
      "loss": 0.0068,
      "step": 4500
    },
    {
      "epoch": 1.04,
      "learning_rate": 2.4109362054681028e-05,
      "loss": 0.0062,
      "step": 5000
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.152029826014913e-05,
      "loss": 0.0042,
      "step": 5500
    },
    {
      "epoch": 1.24,
      "learning_rate": 1.8931234465617236e-05,
      "loss": 0.0043,
      "step": 6000
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.6342170671085334e-05,
      "loss": 0.0054,
      "step": 6500
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.375310687655344e-05,
      "loss": 0.0015,
      "step": 7000
    },
    {
      "epoch": 1.55,
      "learning_rate": 1.1164043082021542e-05,
      "loss": 0.0046,
      "step": 7500
    },
    {
      "epoch": 1.66,
      "learning_rate": 8.574979287489644e-06,
      "loss": 0.0025,
      "step": 8000
    },
    {
      "epoch": 1.76,
      "learning_rate": 5.9859154929577465e-06,
      "loss": 0.0035,
      "step": 8500
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.396851698425849e-06,
      "loss": 0.0016,
      "step": 9000
    },
    {
      "epoch": 1.97,
      "learning_rate": 8.07787903893952e-07,
      "loss": 0.0031,
      "step": 9500
    },
    {
      "epoch": 2.0,
      "step": 9656,
      "total_flos": 2.570302678695936e+16,
      "train_runtime": 1563.26,
      "train_samples_per_second": 6.177
    }
  ],
  "max_steps": 9656,
  "num_train_epochs": 2,
  "total_flos": 2.570302678695936e+16,
  "trial_name": null,
  "trial_params": null
}