{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2213960182297914,
  "eval_steps": 500,
  "global_step": 185,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.9999928710990414e-05,
      "loss": 1.3025,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.9971497912068014e-05,
      "loss": 1.0331,
      "step": 20
    },
    {
      "epoch": 0.1,
      "learning_rate": 1.9886154122075344e-05,
      "loss": 0.9573,
      "step": 40
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.974445512526336e-05,
      "loss": 0.9535,
      "step": 60
    },
    {
      "epoch": 0.19,
      "learning_rate": 1.954720866508546e-05,
      "loss": 0.9354,
      "step": 80
    },
    {
      "epoch": 0.22,
      "eval_loss": 0.9210964441299438,
      "eval_runtime": 48.914,
      "eval_samples_per_second": 19.544,
      "eval_steps_per_second": 9.772,
      "step": 92
    },
    {
      "epoch": 1.02,
      "learning_rate": 1.9295539128732096e-05,
      "loss": 0.8984,
      "step": 100
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.899088113765426e-05,
      "loss": 0.8842,
      "step": 120
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.863497136962213e-05,
      "loss": 0.8736,
      "step": 140
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.8229838658936566e-05,
      "loss": 0.8752,
      "step": 160
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.7777792431226384e-05,
      "loss": 0.8646,
      "step": 180
    },
    {
      "epoch": 1.22,
      "eval_loss": 0.8958887457847595,
      "eval_runtime": 48.9138,
      "eval_samples_per_second": 19.545,
      "eval_steps_per_second": 9.772,
      "step": 185
    },
    {
      "epoch": 1.22,
      "step": 185,
      "total_flos": 9.424742439400243e+16,
      "train_loss": 0.9214719901213775,
      "train_runtime": 5680.6096,
      "train_samples_per_second": 5.871,
      "train_steps_per_second": 0.146
    }
  ],
  "logging_steps": 20,
  "max_steps": 832,
  "num_train_epochs": 2,
  "save_steps": 40,
  "total_flos": 9.424742439400243e+16,
  "trial_name": null,
  "trial_params": null
}