{
  "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": 2.9096331104283392e-05,
      "loss": 1.8932,
      "step": 500
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.819266220856678e-05,
      "loss": 1.2013,
      "step": 1000
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.7288993312850172e-05,
      "loss": 1.1212,
      "step": 1500
    },
    {
      "epoch": 0.36,
      "learning_rate": 2.6385324417133564e-05,
      "loss": 1.0282,
      "step": 2000
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.5481655521416952e-05,
      "loss": 0.9943,
      "step": 2500
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.4577986625700344e-05,
      "loss": 0.9387,
      "step": 3000
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.3674317729983735e-05,
      "loss": 0.8737,
      "step": 3500
    },
    {
      "epoch": 0.72,
      "learning_rate": 2.2770648834267124e-05,
      "loss": 0.9147,
      "step": 4000
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.1866979938550515e-05,
      "loss": 0.9023,
      "step": 4500
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.0963311042833907e-05,
      "loss": 0.8754,
      "step": 5000
    },
    {
      "epoch": 0.99,
      "learning_rate": 2.0059642147117295e-05,
      "loss": 0.8672,
      "step": 5500
    },
    {
      "epoch": 1.08,
      "learning_rate": 1.9155973251400687e-05,
      "loss": 0.6576,
      "step": 6000
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.825230435568408e-05,
      "loss": 0.6301,
      "step": 6500
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.7348635459967467e-05,
      "loss": 0.6397,
      "step": 7000
    },
    {
      "epoch": 1.36,
      "learning_rate": 1.644496656425086e-05,
      "loss": 0.6494,
      "step": 7500
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.554129766853425e-05,
      "loss": 0.6597,
      "step": 8000
    },
    {
      "epoch": 1.54,
      "learning_rate": 1.463762877281764e-05,
      "loss": 0.6255,
      "step": 8500
    },
    {
      "epoch": 1.63,
      "learning_rate": 1.373395987710103e-05,
      "loss": 0.6429,
      "step": 9000
    },
    {
      "epoch": 1.72,
      "learning_rate": 1.283029098138442e-05,
      "loss": 0.6227,
      "step": 9500
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.1926622085667812e-05,
      "loss": 0.6501,
      "step": 10000
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.1022953189951202e-05,
      "loss": 0.622,
      "step": 10500
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.0119284294234592e-05,
      "loss": 0.6297,
      "step": 11000
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.215615398517983e-06,
      "loss": 0.4721,
      "step": 11500
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.311946502801373e-06,
      "loss": 0.4574,
      "step": 12000
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.408277607084765e-06,
      "loss": 0.4401,
      "step": 12500
    },
    {
      "epoch": 2.35,
      "learning_rate": 6.504608711368155e-06,
      "loss": 0.443,
      "step": 13000
    },
    {
      "epoch": 2.44,
      "learning_rate": 5.6009398156515455e-06,
      "loss": 0.4516,
      "step": 13500
    },
    {
      "epoch": 2.53,
      "learning_rate": 4.697270919934936e-06,
      "loss": 0.4649,
      "step": 14000
    },
    {
      "epoch": 2.62,
      "learning_rate": 3.7936020242183263e-06,
      "loss": 0.4485,
      "step": 14500
    },
    {
      "epoch": 2.71,
      "learning_rate": 2.889933128501717e-06,
      "loss": 0.4442,
      "step": 15000
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.9862642327851074e-06,
      "loss": 0.4249,
      "step": 15500
    },
    {
      "epoch": 2.89,
      "learning_rate": 1.0825953370684982e-06,
      "loss": 0.4623,
      "step": 16000
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.7892644135188868e-07,
      "loss": 0.4516,
      "step": 16500
    },
    {
      "epoch": 3.0,
      "step": 16599,
      "total_flos": 5.204482670991974e+16,
      "train_loss": 0.7134747434232642,
      "train_runtime": 6555.6242,
      "train_samples_per_second": 40.511,
      "train_steps_per_second": 2.532
    }
  ],
  "max_steps": 16599,
  "num_train_epochs": 3,
  "total_flos": 5.204482670991974e+16,
  "trial_name": null,
  "trial_params": null
}