{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 17.71217712177122,
  "global_step": 18000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.49,
      "learning_rate": 5e-05,
      "loss": 7.6438,
      "step": 500
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 5.7351,
      "step": 1000
    },
    {
      "epoch": 1.0,
      "eval_loss": 4.940077781677246,
      "eval_runtime": 528.7918,
      "eval_samples_per_second": 134.174,
      "eval_steps_per_second": 4.194,
      "step": 1016
    },
    {
      "epoch": 1.48,
      "learning_rate": 9.705882352941177e-05,
      "loss": 4.4237,
      "step": 1500
    },
    {
      "epoch": 1.97,
      "learning_rate": 9.411764705882353e-05,
      "loss": 3.5768,
      "step": 2000
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.184535503387451,
      "eval_runtime": 528.8582,
      "eval_samples_per_second": 134.157,
      "eval_steps_per_second": 4.194,
      "step": 2032
    },
    {
      "epoch": 2.46,
      "learning_rate": 9.11764705882353e-05,
      "loss": 3.133,
      "step": 2500
    },
    {
      "epoch": 2.95,
      "learning_rate": 8.823529411764706e-05,
      "loss": 2.8394,
      "step": 3000
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.624265432357788,
      "eval_runtime": 528.7795,
      "eval_samples_per_second": 134.177,
      "eval_steps_per_second": 4.195,
      "step": 3048
    },
    {
      "epoch": 3.44,
      "learning_rate": 8.529411764705883e-05,
      "loss": 2.6312,
      "step": 3500
    },
    {
      "epoch": 3.94,
      "learning_rate": 8.23529411764706e-05,
      "loss": 2.4785,
      "step": 4000
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.317702293395996,
      "eval_runtime": 528.7913,
      "eval_samples_per_second": 134.174,
      "eval_steps_per_second": 4.194,
      "step": 4065
    },
    {
      "epoch": 4.43,
      "learning_rate": 7.941176470588235e-05,
      "loss": 2.3492,
      "step": 4500
    },
    {
      "epoch": 4.92,
      "learning_rate": 7.647058823529411e-05,
      "loss": 2.2543,
      "step": 5000
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.1236534118652344,
      "eval_runtime": 528.7093,
      "eval_samples_per_second": 134.195,
      "eval_steps_per_second": 4.195,
      "step": 5081
    },
    {
      "epoch": 5.41,
      "learning_rate": 7.352941176470589e-05,
      "loss": 2.1703,
      "step": 5500
    },
    {
      "epoch": 5.9,
      "learning_rate": 7.058823529411765e-05,
      "loss": 2.0987,
      "step": 6000
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.9984703063964844,
      "eval_runtime": 528.5607,
      "eval_samples_per_second": 134.232,
      "eval_steps_per_second": 4.196,
      "step": 6097
    },
    {
      "epoch": 6.4,
      "learning_rate": 6.764705882352942e-05,
      "loss": 2.0345,
      "step": 6500
    },
    {
      "epoch": 6.89,
      "learning_rate": 6.470588235294118e-05,
      "loss": 1.9782,
      "step": 7000
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.912542462348938,
      "eval_runtime": 528.7872,
      "eval_samples_per_second": 134.175,
      "eval_steps_per_second": 4.195,
      "step": 7113
    },
    {
      "epoch": 7.38,
      "learning_rate": 6.176470588235295e-05,
      "loss": 1.9332,
      "step": 7500
    },
    {
      "epoch": 7.87,
      "learning_rate": 5.882352941176471e-05,
      "loss": 1.8975,
      "step": 8000
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.8240903615951538,
      "eval_runtime": 528.7253,
      "eval_samples_per_second": 134.191,
      "eval_steps_per_second": 4.195,
      "step": 8130
    },
    {
      "epoch": 8.36,
      "learning_rate": 5.588235294117647e-05,
      "loss": 1.8558,
      "step": 8500
    },
    {
      "epoch": 8.86,
      "learning_rate": 5.294117647058824e-05,
      "loss": 1.8258,
      "step": 9000
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.7565855979919434,
      "eval_runtime": 529.8424,
      "eval_samples_per_second": 133.908,
      "eval_steps_per_second": 4.186,
      "step": 9146
    },
    {
      "epoch": 9.35,
      "learning_rate": 5e-05,
      "loss": 1.7841,
      "step": 9500
    },
    {
      "epoch": 9.84,
      "learning_rate": 4.705882352941177e-05,
      "loss": 1.7621,
      "step": 10000
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.710971713066101,
      "eval_runtime": 529.8142,
      "eval_samples_per_second": 133.915,
      "eval_steps_per_second": 4.186,
      "step": 10162
    },
    {
      "epoch": 10.33,
      "learning_rate": 4.411764705882353e-05,
      "loss": 1.7383,
      "step": 10500
    },
    {
      "epoch": 10.82,
      "learning_rate": 4.11764705882353e-05,
      "loss": 1.711,
      "step": 11000
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.6690000295639038,
      "eval_runtime": 528.7385,
      "eval_samples_per_second": 134.187,
      "eval_steps_per_second": 4.195,
      "step": 11178
    },
    {
      "epoch": 11.32,
      "learning_rate": 3.8235294117647055e-05,
      "loss": 1.6908,
      "step": 11500
    },
    {
      "epoch": 11.81,
      "learning_rate": 3.529411764705883e-05,
      "loss": 1.6665,
      "step": 12000
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.6373904943466187,
      "eval_runtime": 528.848,
      "eval_samples_per_second": 134.16,
      "eval_steps_per_second": 4.194,
      "step": 12195
    },
    {
      "epoch": 12.3,
      "learning_rate": 3.235294117647059e-05,
      "loss": 1.6546,
      "step": 12500
    },
    {
      "epoch": 12.79,
      "learning_rate": 2.9411764705882354e-05,
      "loss": 1.6246,
      "step": 13000
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.5987190008163452,
      "eval_runtime": 528.7748,
      "eval_samples_per_second": 134.178,
      "eval_steps_per_second": 4.195,
      "step": 13211
    },
    {
      "epoch": 13.28,
      "learning_rate": 2.647058823529412e-05,
      "loss": 1.6166,
      "step": 13500
    },
    {
      "epoch": 13.78,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 1.603,
      "step": 14000
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.5721200704574585,
      "eval_runtime": 528.6668,
      "eval_samples_per_second": 134.206,
      "eval_steps_per_second": 4.195,
      "step": 14227
    },
    {
      "epoch": 14.27,
      "learning_rate": 2.058823529411765e-05,
      "loss": 1.5918,
      "step": 14500
    },
    {
      "epoch": 14.76,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 1.5762,
      "step": 15000
    },
    {
      "epoch": 15.0,
      "eval_loss": 1.5530680418014526,
      "eval_runtime": 528.7404,
      "eval_samples_per_second": 134.187,
      "eval_steps_per_second": 4.195,
      "step": 15243
    },
    {
      "epoch": 15.25,
      "learning_rate": 1.4705882352941177e-05,
      "loss": 1.5625,
      "step": 15500
    },
    {
      "epoch": 15.74,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.5572,
      "step": 16000
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.535999059677124,
      "eval_runtime": 528.8228,
      "eval_samples_per_second": 134.166,
      "eval_steps_per_second": 4.194,
      "step": 16260
    },
    {
      "epoch": 16.24,
      "learning_rate": 8.823529411764707e-06,
      "loss": 1.5418,
      "step": 16500
    },
    {
      "epoch": 16.73,
      "learning_rate": 5.882352941176471e-06,
      "loss": 1.5338,
      "step": 17000
    },
    {
      "epoch": 17.0,
      "eval_loss": 1.535353422164917,
      "eval_runtime": 528.5026,
      "eval_samples_per_second": 134.247,
      "eval_steps_per_second": 4.197,
      "step": 17276
    },
    {
      "epoch": 17.22,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 1.5272,
      "step": 17500
    },
    {
      "epoch": 17.71,
      "learning_rate": 0.0,
      "loss": 1.524,
      "step": 18000
    }
  ],
  "max_steps": 18000,
  "num_train_epochs": 18,
  "total_flos": 4.738018441728768e+17,
  "trial_name": null,
  "trial_params": null
}