{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 129,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "grad_norm": 8701.357710303886,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 28.3212,
      "step": 1
    },
    {
      "epoch": 0.12,
      "grad_norm": 481.362737716594,
      "learning_rate": 7.692307692307694e-06,
      "loss": 22.1134,
      "step": 5
    },
    {
      "epoch": 0.23,
      "grad_norm": 259.4834207727077,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 14.8237,
      "step": 10
    },
    {
      "epoch": 0.35,
      "grad_norm": 131.82252822714617,
      "learning_rate": 1.998533413851124e-05,
      "loss": 12.5705,
      "step": 15
    },
    {
      "epoch": 0.47,
      "grad_norm": 387.9499566391837,
      "learning_rate": 1.982083682742156e-05,
      "loss": 13.7286,
      "step": 20
    },
    {
      "epoch": 0.58,
      "grad_norm": 219.00627726463617,
      "learning_rate": 1.9476531711828027e-05,
      "loss": 11.3422,
      "step": 25
    },
    {
      "epoch": 0.7,
      "grad_norm": 221.35867761333137,
      "learning_rate": 1.8958722607586883e-05,
      "loss": 7.7604,
      "step": 30
    },
    {
      "epoch": 0.81,
      "grad_norm": 43.14746736074378,
      "learning_rate": 1.827688998156891e-05,
      "loss": 3.4082,
      "step": 35
    },
    {
      "epoch": 0.93,
      "grad_norm": 29.64836419071723,
      "learning_rate": 1.7443517375622706e-05,
      "loss": 2.9518,
      "step": 40
    },
    {
      "epoch": 1.05,
      "grad_norm": 21.427062393117673,
      "learning_rate": 1.647386284781828e-05,
      "loss": 2.6311,
      "step": 45
    },
    {
      "epoch": 1.16,
      "grad_norm": 19.158148236389312,
      "learning_rate": 1.5385679615609045e-05,
      "loss": 2.4838,
      "step": 50
    },
    {
      "epoch": 1.28,
      "grad_norm": 14.070587316942852,
      "learning_rate": 1.4198891015602648e-05,
      "loss": 2.3473,
      "step": 55
    },
    {
      "epoch": 1.4,
      "grad_norm": 34.76326434361083,
      "learning_rate": 1.2935225731039349e-05,
      "loss": 2.2796,
      "step": 60
    },
    {
      "epoch": 1.51,
      "grad_norm": 16.17405801850995,
      "learning_rate": 1.161781996552765e-05,
      "loss": 2.1971,
      "step": 65
    },
    {
      "epoch": 1.63,
      "grad_norm": 10.06558936803853,
      "learning_rate": 1.0270793846761347e-05,
      "loss": 2.1598,
      "step": 70
    },
    {
      "epoch": 1.74,
      "grad_norm": 21.625955834057603,
      "learning_rate": 8.918809815760585e-06,
      "loss": 2.148,
      "step": 75
    },
    {
      "epoch": 1.86,
      "grad_norm": 10.59659968673671,
      "learning_rate": 7.586621087002945e-06,
      "loss": 2.0663,
      "step": 80
    },
    {
      "epoch": 1.98,
      "grad_norm": 10.684749100687451,
      "learning_rate": 6.298618446600856e-06,
      "loss": 2.079,
      "step": 85
    },
    {
      "epoch": 2.09,
      "grad_norm": 18.49345747302597,
      "learning_rate": 5.078383686109927e-06,
      "loss": 2.0018,
      "step": 90
    },
    {
      "epoch": 2.21,
      "grad_norm": 14.92738780630304,
      "learning_rate": 3.948257848062351e-06,
      "loss": 1.9669,
      "step": 95
    },
    {
      "epoch": 2.33,
      "grad_norm": 9.352817552145108,
      "learning_rate": 2.9289321881345257e-06,
      "loss": 1.9691,
      "step": 100
    },
    {
      "epoch": 2.44,
      "grad_norm": 7.228395436940449,
      "learning_rate": 2.0390693429435626e-06,
      "loss": 1.9298,
      "step": 105
    },
    {
      "epoch": 2.56,
      "grad_norm": 7.290178837796146,
      "learning_rate": 1.2949616394382802e-06,
      "loss": 1.9519,
      "step": 110
    },
    {
      "epoch": 2.67,
      "grad_norm": 6.909179219835438,
      "learning_rate": 7.102328018320859e-07,
      "loss": 1.9449,
      "step": 115
    },
    {
      "epoch": 2.79,
      "grad_norm": 6.799236287899626,
      "learning_rate": 2.955885174678852e-07,
      "loss": 1.9203,
      "step": 120
    },
    {
      "epoch": 2.91,
      "grad_norm": 7.120567007446623,
      "learning_rate": 5.862042845640403e-08,
      "loss": 1.9294,
      "step": 125
    },
    {
      "epoch": 3.0,
      "step": 129,
      "total_flos": 8882126585856.0,
      "train_loss": 4.938086860863737,
      "train_runtime": 252.0914,
      "train_samples_per_second": 16.232,
      "train_steps_per_second": 0.512
    }
  ],
  "logging_steps": 5,
  "max_steps": 129,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 8882126585856.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}