{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7171029042667623,
  "eval_steps": 2000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 2.5916264057159424,
      "learning_rate": 1e-06,
      "loss": 1.0776,
      "step": 100
    },
    {
      "epoch": 0.07,
      "grad_norm": 3.6381211280822754,
      "learning_rate": 9.898989898989898e-07,
      "loss": 0.8345,
      "step": 200
    },
    {
      "epoch": 0.11,
      "grad_norm": 3.9932849407196045,
      "learning_rate": 9.7989898989899e-07,
      "loss": 0.6863,
      "step": 300
    },
    {
      "epoch": 0.14,
      "grad_norm": 2.8963606357574463,
      "learning_rate": 9.697979797979798e-07,
      "loss": 0.6659,
      "step": 400
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.000125885009766,
      "learning_rate": 9.598989898989899e-07,
      "loss": 0.6552,
      "step": 500
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.7767927646636963,
      "learning_rate": 9.497979797979798e-07,
      "loss": 0.6583,
      "step": 600
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.204336166381836,
      "learning_rate": 9.396969696969696e-07,
      "loss": 0.6229,
      "step": 700
    },
    {
      "epoch": 0.29,
      "grad_norm": 3.798582077026367,
      "learning_rate": 9.295959595959596e-07,
      "loss": 0.6138,
      "step": 800
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.92914080619812,
      "learning_rate": 9.194949494949495e-07,
      "loss": 0.6266,
      "step": 900
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.982314109802246,
      "learning_rate": 9.093939393939394e-07,
      "loss": 0.6311,
      "step": 1000
    },
    {
      "epoch": 0.39,
      "grad_norm": 3.385096311569214,
      "learning_rate": 8.992929292929292e-07,
      "loss": 0.6141,
      "step": 1100
    },
    {
      "epoch": 0.43,
      "grad_norm": 3.015211343765259,
      "learning_rate": 8.891919191919191e-07,
      "loss": 0.6369,
      "step": 1200
    },
    {
      "epoch": 0.47,
      "grad_norm": 5.285658359527588,
      "learning_rate": 8.790909090909091e-07,
      "loss": 0.6442,
      "step": 1300
    },
    {
      "epoch": 0.5,
      "grad_norm": 4.69854736328125,
      "learning_rate": 8.68989898989899e-07,
      "loss": 0.6419,
      "step": 1400
    },
    {
      "epoch": 0.54,
      "grad_norm": 2.95841121673584,
      "learning_rate": 8.58989898989899e-07,
      "loss": 0.6575,
      "step": 1500
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.4997267723083496,
      "learning_rate": 8.488888888888888e-07,
      "loss": 0.6346,
      "step": 1600
    },
    {
      "epoch": 0.61,
      "grad_norm": 4.126640319824219,
      "learning_rate": 8.387878787878787e-07,
      "loss": 0.6005,
      "step": 1700
    },
    {
      "epoch": 0.65,
      "grad_norm": 3.040174961090088,
      "learning_rate": 8.286868686868687e-07,
      "loss": 0.6434,
      "step": 1800
    },
    {
      "epoch": 0.68,
      "grad_norm": 4.870083332061768,
      "learning_rate": 8.185858585858586e-07,
      "loss": 0.6202,
      "step": 1900
    },
    {
      "epoch": 0.72,
      "grad_norm": 4.307583808898926,
      "learning_rate": 8.084848484848484e-07,
      "loss": 0.6552,
      "step": 2000
    },
    {
      "epoch": 0.72,
      "eval_loss": 0.746256411075592,
      "eval_runtime": 199.9457,
      "eval_samples_per_second": 5.001,
      "eval_steps_per_second": 1.25,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 2000,
  "total_flos": 1.88564197343232e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}