{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100.0,
  "global_step": 1176,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "grad_norm": 20.62335205078125,
      "learning_rate": 4.5748299319727895e-05,
      "loss": 3.0207,
      "step": 100
    },
    {
      "epoch": 0.51,
      "grad_norm": 24.41301155090332,
      "learning_rate": 4.149659863945579e-05,
      "loss": 2.8238,
      "step": 200
    },
    {
      "epoch": 0.77,
      "grad_norm": 23.851058959960938,
      "learning_rate": 3.724489795918368e-05,
      "loss": 2.7176,
      "step": 300
    },
    {
      "epoch": 1.02,
      "grad_norm": 21.198942184448242,
      "learning_rate": 3.2993197278911564e-05,
      "loss": 2.606,
      "step": 400
    },
    {
      "epoch": 1.28,
      "grad_norm": 22.139713287353516,
      "learning_rate": 2.8741496598639456e-05,
      "loss": 2.1191,
      "step": 500
    },
    {
      "epoch": 1.53,
      "grad_norm": 26.165542602539062,
      "learning_rate": 2.448979591836735e-05,
      "loss": 2.1034,
      "step": 600
    },
    {
      "epoch": 1.79,
      "grad_norm": 30.156545639038086,
      "learning_rate": 2.023809523809524e-05,
      "loss": 2.0801,
      "step": 700
    },
    {
      "epoch": 2.04,
      "grad_norm": 16.38189697265625,
      "learning_rate": 1.5986394557823133e-05,
      "loss": 1.908,
      "step": 800
    },
    {
      "epoch": 2.3,
      "grad_norm": 21.190061569213867,
      "learning_rate": 1.1734693877551021e-05,
      "loss": 1.4389,
      "step": 900
    },
    {
      "epoch": 2.55,
      "grad_norm": 28.82680320739746,
      "learning_rate": 7.482993197278912e-06,
      "loss": 1.4339,
      "step": 1000
    },
    {
      "epoch": 2.81,
      "grad_norm": 26.55443572998047,
      "learning_rate": 3.231292517006803e-06,
      "loss": 1.3741,
      "step": 1100
    },
    {
      "epoch": 3.0,
      "step": 1176,
      "total_flos": 4376310425967942.0,
      "train_loss": 2.0960015277473296,
      "train_runtime": 1329.2473,
      "train_samples_per_second": 56.601,
      "train_steps_per_second": 0.885
    }
  ],
  "logging_steps": 100,
  "max_steps": 1176,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 4376310425967942.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}