{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.0,
  "global_step": 11268,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.27,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 5.9906,
      "step": 500
    },
    {
      "epoch": 0.53,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 5.31,
      "step": 1000
    },
    {
      "epoch": 0.8,
      "learning_rate": 5e-05,
      "loss": 5.0291,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_loss": 4.711434364318848,
      "eval_runtime": 48.7776,
      "eval_samples_per_second": 34.237,
      "eval_steps_per_second": 4.285,
      "step": 1878
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.744062244062245e-05,
      "loss": 4.8125,
      "step": 2000
    },
    {
      "epoch": 1.33,
      "learning_rate": 4.4881244881244886e-05,
      "loss": 4.4719,
      "step": 2500
    },
    {
      "epoch": 1.6,
      "learning_rate": 4.2321867321867324e-05,
      "loss": 4.4173,
      "step": 3000
    },
    {
      "epoch": 1.86,
      "learning_rate": 3.976248976248976e-05,
      "loss": 4.3799,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_loss": 4.4446024894714355,
      "eval_runtime": 48.934,
      "eval_samples_per_second": 34.128,
      "eval_steps_per_second": 4.271,
      "step": 3756
    },
    {
      "epoch": 2.13,
      "learning_rate": 3.720311220311221e-05,
      "loss": 4.0749,
      "step": 4000
    },
    {
      "epoch": 2.4,
      "learning_rate": 3.4643734643734647e-05,
      "loss": 3.8423,
      "step": 4500
    },
    {
      "epoch": 2.66,
      "learning_rate": 3.2084357084357085e-05,
      "loss": 3.8255,
      "step": 5000
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.952497952497953e-05,
      "loss": 3.8095,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_loss": 4.39353084564209,
      "eval_runtime": 48.7824,
      "eval_samples_per_second": 34.234,
      "eval_steps_per_second": 4.284,
      "step": 5634
    },
    {
      "epoch": 3.19,
      "learning_rate": 2.6965601965601965e-05,
      "loss": 3.4428,
      "step": 6000
    },
    {
      "epoch": 3.46,
      "learning_rate": 2.4406224406224407e-05,
      "loss": 3.2758,
      "step": 6500
    },
    {
      "epoch": 3.73,
      "learning_rate": 2.1846846846846845e-05,
      "loss": 3.2939,
      "step": 7000
    },
    {
      "epoch": 3.99,
      "learning_rate": 1.928746928746929e-05,
      "loss": 3.2732,
      "step": 7500
    },
    {
      "epoch": 4.0,
      "eval_loss": 4.4684529304504395,
      "eval_runtime": 48.7667,
      "eval_samples_per_second": 34.245,
      "eval_steps_per_second": 4.286,
      "step": 7512
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.672809172809173e-05,
      "loss": 2.8454,
      "step": 8000
    },
    {
      "epoch": 4.53,
      "learning_rate": 1.416871416871417e-05,
      "loss": 2.8636,
      "step": 8500
    },
    {
      "epoch": 4.79,
      "learning_rate": 1.1609336609336611e-05,
      "loss": 2.8471,
      "step": 9000
    },
    {
      "epoch": 5.0,
      "eval_loss": 4.63837194442749,
      "eval_runtime": 48.8194,
      "eval_samples_per_second": 34.208,
      "eval_steps_per_second": 4.281,
      "step": 9390
    },
    {
      "epoch": 5.06,
      "learning_rate": 9.049959049959051e-06,
      "loss": 2.7907,
      "step": 9500
    },
    {
      "epoch": 5.32,
      "learning_rate": 6.49058149058149e-06,
      "loss": 2.5218,
      "step": 10000
    },
    {
      "epoch": 5.59,
      "learning_rate": 3.931203931203931e-06,
      "loss": 2.5563,
      "step": 10500
    },
    {
      "epoch": 5.86,
      "learning_rate": 1.371826371826372e-06,
      "loss": 2.5324,
      "step": 11000
    },
    {
      "epoch": 6.0,
      "eval_loss": 4.8097243309021,
      "eval_runtime": 48.8246,
      "eval_samples_per_second": 34.204,
      "eval_steps_per_second": 4.281,
      "step": 11268
    }
  ],
  "max_steps": 11268,
  "num_train_epochs": 6,
  "total_flos": 5.512311147331584e+16,
  "trial_name": null,
  "trial_params": null
}