{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5252100840336135,
  "eval_steps": 500,
  "global_step": 250,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.052521008403361345,
      "grad_norm": 8.114081382751465,
      "learning_rate": 4.994197170276449e-05,
      "loss": 2.6235,
      "step": 25
    },
    {
      "epoch": 0.10504201680672269,
      "grad_norm": 7.403553009033203,
      "learning_rate": 4.929224310196645e-05,
      "loss": 2.7435,
      "step": 50
    },
    {
      "epoch": 0.15756302521008403,
      "grad_norm": 6.665265083312988,
      "learning_rate": 4.7939126731158055e-05,
      "loss": 2.7157,
      "step": 75
    },
    {
      "epoch": 0.21008403361344538,
      "grad_norm": 377.5302734375,
      "learning_rate": 4.592180240187963e-05,
      "loss": 2.8261,
      "step": 100
    },
    {
      "epoch": 0.26260504201680673,
      "grad_norm": 20.072336196899414,
      "learning_rate": 4.329868222763601e-05,
      "loss": 2.6462,
      "step": 125
    },
    {
      "epoch": 0.31512605042016806,
      "grad_norm": 6.688249111175537,
      "learning_rate": 4.014571928703281e-05,
      "loss": 2.5808,
      "step": 150
    },
    {
      "epoch": 0.36764705882352944,
      "grad_norm": 5.843264102935791,
      "learning_rate": 3.6554208384028955e-05,
      "loss": 2.8584,
      "step": 175
    },
    {
      "epoch": 0.42016806722689076,
      "grad_norm": 5.064057350158691,
      "learning_rate": 3.262814258482371e-05,
      "loss": 2.423,
      "step": 200
    },
    {
      "epoch": 0.4726890756302521,
      "grad_norm": 6.219644546508789,
      "learning_rate": 2.8481202073491685e-05,
      "loss": 2.3885,
      "step": 225
    },
    {
      "epoch": 0.5252100840336135,
      "grad_norm": 10.170488357543945,
      "learning_rate": 2.4233462514778545e-05,
      "loss": 2.2676,
      "step": 250
    }
  ],
  "logging_steps": 25,
  "max_steps": 476,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 250,
  "total_flos": 2.3055128592384e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}