{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 300.0,
  "global_step": 16500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.09,
      "learning_rate": 1.9393939393939395e-05,
      "loss": 0.0325,
      "step": 500
    },
    {
      "epoch": 18.18,
      "learning_rate": 1.8787878787878792e-05,
      "loss": 0.0205,
      "step": 1000
    },
    {
      "epoch": 18.18,
      "eval_loss": 0.024596277624368668,
      "eval_runtime": 0.6803,
      "eval_samples_per_second": 385.138,
      "eval_steps_per_second": 24.99,
      "step": 1000
    },
    {
      "epoch": 27.27,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.0176,
      "step": 1500
    },
    {
      "epoch": 36.36,
      "learning_rate": 1.7575757575757576e-05,
      "loss": 0.0148,
      "step": 2000
    },
    {
      "epoch": 36.36,
      "eval_loss": 0.009050876833498478,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 385.757,
      "eval_steps_per_second": 25.03,
      "step": 2000
    },
    {
      "epoch": 45.45,
      "learning_rate": 1.6969696969696972e-05,
      "loss": 0.0139,
      "step": 2500
    },
    {
      "epoch": 54.55,
      "learning_rate": 1.6363636363636366e-05,
      "loss": 0.0131,
      "step": 3000
    },
    {
      "epoch": 54.55,
      "eval_loss": 0.08356618881225586,
      "eval_runtime": 0.6799,
      "eval_samples_per_second": 385.336,
      "eval_steps_per_second": 25.003,
      "step": 3000
    },
    {
      "epoch": 63.64,
      "learning_rate": 1.575757575757576e-05,
      "loss": 0.0123,
      "step": 3500
    },
    {
      "epoch": 72.73,
      "learning_rate": 1.5151515151515153e-05,
      "loss": 0.012,
      "step": 4000
    },
    {
      "epoch": 72.73,
      "eval_loss": 0.0886448472738266,
      "eval_runtime": 0.6927,
      "eval_samples_per_second": 378.228,
      "eval_steps_per_second": 24.542,
      "step": 4000
    },
    {
      "epoch": 81.82,
      "learning_rate": 1.4545454545454546e-05,
      "loss": 0.0114,
      "step": 4500
    },
    {
      "epoch": 90.91,
      "learning_rate": 1.3939393939393942e-05,
      "loss": 0.011,
      "step": 5000
    },
    {
      "epoch": 90.91,
      "eval_loss": 0.0763353630900383,
      "eval_runtime": 0.678,
      "eval_samples_per_second": 386.405,
      "eval_steps_per_second": 25.072,
      "step": 5000
    },
    {
      "epoch": 100.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.0105,
      "step": 5500
    },
    {
      "epoch": 109.09,
      "learning_rate": 1.2727272727272728e-05,
      "loss": 0.0111,
      "step": 6000
    },
    {
      "epoch": 109.09,
      "eval_loss": 0.10684655606746674,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 385.762,
      "eval_steps_per_second": 25.03,
      "step": 6000
    },
    {
      "epoch": 118.18,
      "learning_rate": 1.2121212121212122e-05,
      "loss": 0.0104,
      "step": 6500
    },
    {
      "epoch": 127.27,
      "learning_rate": 1.1515151515151517e-05,
      "loss": 0.0104,
      "step": 7000
    },
    {
      "epoch": 127.27,
      "eval_loss": 0.0989227443933487,
      "eval_runtime": 0.6773,
      "eval_samples_per_second": 386.802,
      "eval_steps_per_second": 25.098,
      "step": 7000
    },
    {
      "epoch": 136.36,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 0.0097,
      "step": 7500
    },
    {
      "epoch": 145.45,
      "learning_rate": 1.0303030303030304e-05,
      "loss": 0.0106,
      "step": 8000
    },
    {
      "epoch": 145.45,
      "eval_loss": 0.09557081758975983,
      "eval_runtime": 0.7034,
      "eval_samples_per_second": 372.496,
      "eval_steps_per_second": 24.17,
      "step": 8000
    },
    {
      "epoch": 154.55,
      "learning_rate": 9.696969696969698e-06,
      "loss": 0.0099,
      "step": 8500
    },
    {
      "epoch": 163.64,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.0097,
      "step": 9000
    },
    {
      "epoch": 163.64,
      "eval_loss": 0.07463082671165466,
      "eval_runtime": 0.6685,
      "eval_samples_per_second": 391.9,
      "eval_steps_per_second": 25.429,
      "step": 9000
    },
    {
      "epoch": 172.73,
      "learning_rate": 8.484848484848486e-06,
      "loss": 0.0111,
      "step": 9500
    },
    {
      "epoch": 181.82,
      "learning_rate": 7.87878787878788e-06,
      "loss": 0.0097,
      "step": 10000
    },
    {
      "epoch": 181.82,
      "eval_loss": 0.0682496652007103,
      "eval_runtime": 0.6795,
      "eval_samples_per_second": 385.591,
      "eval_steps_per_second": 25.019,
      "step": 10000
    },
    {
      "epoch": 190.91,
      "learning_rate": 7.272727272727273e-06,
      "loss": 0.0098,
      "step": 10500
    },
    {
      "epoch": 200.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.0095,
      "step": 11000
    },
    {
      "epoch": 200.0,
      "eval_loss": 0.04372559115290642,
      "eval_runtime": 0.7607,
      "eval_samples_per_second": 344.421,
      "eval_steps_per_second": 22.348,
      "step": 11000
    },
    {
      "epoch": 209.09,
      "learning_rate": 6.060606060606061e-06,
      "loss": 0.0096,
      "step": 11500
    },
    {
      "epoch": 218.18,
      "learning_rate": 5.4545454545454545e-06,
      "loss": 0.0095,
      "step": 12000
    },
    {
      "epoch": 218.18,
      "eval_loss": 0.0709039568901062,
      "eval_runtime": 0.7061,
      "eval_samples_per_second": 371.075,
      "eval_steps_per_second": 24.077,
      "step": 12000
    },
    {
      "epoch": 227.27,
      "learning_rate": 4.848484848484849e-06,
      "loss": 0.0092,
      "step": 12500
    },
    {
      "epoch": 236.36,
      "learning_rate": 4.242424242424243e-06,
      "loss": 0.0093,
      "step": 13000
    },
    {
      "epoch": 236.36,
      "eval_loss": 0.06207394599914551,
      "eval_runtime": 0.7614,
      "eval_samples_per_second": 344.108,
      "eval_steps_per_second": 22.328,
      "step": 13000
    },
    {
      "epoch": 245.45,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 0.0092,
      "step": 13500
    },
    {
      "epoch": 254.55,
      "learning_rate": 3.0303030303030305e-06,
      "loss": 0.0095,
      "step": 14000
    },
    {
      "epoch": 254.55,
      "eval_loss": 0.0763927549123764,
      "eval_runtime": 0.7074,
      "eval_samples_per_second": 370.363,
      "eval_steps_per_second": 24.031,
      "step": 14000
    },
    {
      "epoch": 263.64,
      "learning_rate": 2.4242424242424244e-06,
      "loss": 0.0094,
      "step": 14500
    },
    {
      "epoch": 272.73,
      "learning_rate": 1.8181818181818183e-06,
      "loss": 0.0095,
      "step": 15000
    },
    {
      "epoch": 272.73,
      "eval_loss": 0.0721823051571846,
      "eval_runtime": 0.6582,
      "eval_samples_per_second": 398.054,
      "eval_steps_per_second": 25.828,
      "step": 15000
    },
    {
      "epoch": 281.82,
      "learning_rate": 1.2121212121212122e-06,
      "loss": 0.0091,
      "step": 15500
    },
    {
      "epoch": 290.91,
      "learning_rate": 6.060606060606061e-07,
      "loss": 0.0092,
      "step": 16000
    },
    {
      "epoch": 290.91,
      "eval_loss": 0.05487630143761635,
      "eval_runtime": 0.715,
      "eval_samples_per_second": 366.453,
      "eval_steps_per_second": 23.777,
      "step": 16000
    },
    {
      "epoch": 300.0,
      "learning_rate": 0.0,
      "loss": 0.0094,
      "step": 16500
    }
  ],
  "max_steps": 16500,
  "num_train_epochs": 300,
  "total_flos": 9.41486238830592e+16,
  "trial_name": null,
  "trial_params": null
}