{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1,
  "eval_steps": 2000,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 0.9697519540786743,
      "learning_rate": 1e-06,
      "loss": 0.1629,
      "step": 100
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3848124742507935,
      "learning_rate": 9.898989898989898e-07,
      "loss": 0.14,
      "step": 200
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.9986572265625,
      "learning_rate": 9.797979797979797e-07,
      "loss": 0.1354,
      "step": 300
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.11438798904418945,
      "learning_rate": 9.696969696969698e-07,
      "loss": 0.1182,
      "step": 400
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.8548241257667542,
      "learning_rate": 9.595959595959596e-07,
      "loss": 0.1192,
      "step": 500
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.5312464237213135,
      "learning_rate": 9.494949494949495e-07,
      "loss": 0.0997,
      "step": 600
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.9692059755325317,
      "learning_rate": 9.393939393939395e-07,
      "loss": 0.102,
      "step": 700
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.42864611744880676,
      "learning_rate": 9.292929292929292e-07,
      "loss": 0.0901,
      "step": 800
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.852543830871582,
      "learning_rate": 9.191919191919192e-07,
      "loss": 0.0937,
      "step": 900
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.5718303322792053,
      "learning_rate": 9.09090909090909e-07,
      "loss": 0.093,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.9396565556526184,
      "learning_rate": 8.98989898989899e-07,
      "loss": 0.0892,
      "step": 1100
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.08157779276371002,
      "learning_rate": 8.888888888888888e-07,
      "loss": 0.0934,
      "step": 1200
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.8076322078704834,
      "learning_rate": 8.787878787878787e-07,
      "loss": 0.0725,
      "step": 1300
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.5076119899749756,
      "learning_rate": 8.686868686868687e-07,
      "loss": 0.0835,
      "step": 1400
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.1567238569259644,
      "learning_rate": 8.585858585858586e-07,
      "loss": 0.0747,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.6817927956581116,
      "learning_rate": 8.484848484848484e-07,
      "loss": 0.0903,
      "step": 1600
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.6467050313949585,
      "learning_rate": 8.383838383838383e-07,
      "loss": 0.0721,
      "step": 1700
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.8435570001602173,
      "learning_rate": 8.282828282828283e-07,
      "loss": 0.0847,
      "step": 1800
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.6265794634819031,
      "learning_rate": 8.181818181818182e-07,
      "loss": 0.0687,
      "step": 1900
    },
    {
      "epoch": 0.1,
      "grad_norm": 1.360060453414917,
      "learning_rate": 8.08080808080808e-07,
      "loss": 0.0748,
      "step": 2000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.06745574623346329,
      "eval_runtime": 304.7718,
      "eval_samples_per_second": 3.281,
      "eval_steps_per_second": 0.82,
      "step": 2000
    }
  ],
  "logging_steps": 100,
  "max_steps": 10000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "total_flos": 1.63205502468096e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}