{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1254075746175069,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01254075746175069,
      "grad_norm": 83.43741607666016,
      "learning_rate": 1.4041058439929773e-05,
      "loss": 1.9161,
      "step": 50
    },
    {
      "epoch": 0.02508151492350138,
      "grad_norm": 61.7346305847168,
      "learning_rate": 1.3982116879859543e-05,
      "loss": 2.0225,
      "step": 100
    },
    {
      "epoch": 0.03762227238525207,
      "grad_norm": 59.338584899902344,
      "learning_rate": 1.3923175319789316e-05,
      "loss": 1.8901,
      "step": 150
    },
    {
      "epoch": 0.05016302984700276,
      "grad_norm": 11.058795928955078,
      "learning_rate": 1.3864233759719087e-05,
      "loss": 1.1187,
      "step": 200
    },
    {
      "epoch": 0.06270378730875345,
      "grad_norm": 34.20891571044922,
      "learning_rate": 1.380529219964886e-05,
      "loss": 1.4935,
      "step": 250
    },
    {
      "epoch": 0.07524454477050414,
      "grad_norm": 77.22309875488281,
      "learning_rate": 1.374635063957863e-05,
      "loss": 0.9285,
      "step": 300
    },
    {
      "epoch": 0.08778530223225482,
      "grad_norm": 2.8750481605529785,
      "learning_rate": 1.3687409079508403e-05,
      "loss": 1.1021,
      "step": 350
    },
    {
      "epoch": 0.10032605969400551,
      "grad_norm": 18.73400115966797,
      "learning_rate": 1.3628467519438174e-05,
      "loss": 1.0036,
      "step": 400
    },
    {
      "epoch": 0.11286681715575621,
      "grad_norm": 35.63496780395508,
      "learning_rate": 1.3569525959367947e-05,
      "loss": 1.1348,
      "step": 450
    },
    {
      "epoch": 0.1254075746175069,
      "grad_norm": 45.8936653137207,
      "learning_rate": 1.3510584399297718e-05,
      "loss": 0.826,
      "step": 500
    }
  ],
  "logging_steps": 50,
  "max_steps": 11961,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}