{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.64,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "alignment_loss": 0.957948,
      "epoch": 0.032,
      "grad_norm": 13216.34765625,
      "learning_rate": 9e-06,
      "loss": 604.8516,
      "mean_token_accuracy": 0.6210215302184224,
      "num_tokens": 988428.0,
      "step": 10,
      "teacher_ce_loss": 39.105175
    },
    {
      "alignment_loss": 0.946621,
      "epoch": 0.064,
      "grad_norm": 3091.22509765625,
      "learning_rate": 9.853896103896104e-06,
      "loss": 171.2203,
      "mean_token_accuracy": 0.7685627818107605,
      "num_tokens": 1972507.0,
      "step": 20,
      "teacher_ce_loss": 10.471479
    },
    {
      "alignment_loss": 0.94924,
      "epoch": 0.096,
      "grad_norm": 3218.343017578125,
      "learning_rate": 9.691558441558442e-06,
      "loss": 115.0068,
      "mean_token_accuracy": 0.7847720004618168,
      "num_tokens": 2964436.0,
      "step": 30,
      "teacher_ce_loss": 7.006755
    },
    {
      "alignment_loss": 0.938964,
      "epoch": 0.128,
      "grad_norm": 2620.491455078125,
      "learning_rate": 9.52922077922078e-06,
      "loss": 83.4615,
      "mean_token_accuracy": 0.8658171582967042,
      "num_tokens": 3939005.0,
      "step": 40,
      "teacher_ce_loss": 5.536875
    },
    {
      "alignment_loss": 0.9248,
      "epoch": 0.16,
      "grad_norm": 1182.9658203125,
      "learning_rate": 9.366883116883117e-06,
      "loss": 56.0003,
      "mean_token_accuracy": 0.8695644401013851,
      "num_tokens": 4920499.0,
      "step": 50,
      "teacher_ce_loss": 3.622461
    },
    {
      "alignment_loss": 0.910064,
      "epoch": 0.192,
      "grad_norm": 556.953369140625,
      "learning_rate": 9.204545454545455e-06,
      "loss": 46.6464,
      "mean_token_accuracy": 0.8793487269431353,
      "num_tokens": 5904564.0,
      "step": 60,
      "teacher_ce_loss": 3.133064
    },
    {
      "alignment_loss": 0.896394,
      "epoch": 0.224,
      "grad_norm": 573.4602661132812,
      "learning_rate": 9.042207792207793e-06,
      "loss": 45.5978,
      "mean_token_accuracy": 0.8741835005581379,
      "num_tokens": 6882920.0,
      "step": 70,
      "teacher_ce_loss": 2.513722
    },
    {
      "alignment_loss": 0.885299,
      "epoch": 0.256,
      "grad_norm": 481.3504333496094,
      "learning_rate": 8.879870129870131e-06,
      "loss": 43.9065,
      "mean_token_accuracy": 0.8817355066537857,
      "num_tokens": 7882978.0,
      "step": 80,
      "teacher_ce_loss": 2.451343
    },
    {
      "alignment_loss": 0.872261,
      "epoch": 0.288,
      "grad_norm": 567.1941528320312,
      "learning_rate": 8.717532467532468e-06,
      "loss": 42.6578,
      "mean_token_accuracy": 0.884648934751749,
      "num_tokens": 8868924.0,
      "step": 90,
      "teacher_ce_loss": 2.764072
    },
    {
      "alignment_loss": 0.866291,
      "epoch": 0.32,
      "grad_norm": 437.5704345703125,
      "learning_rate": 8.555194805194806e-06,
      "loss": 41.3949,
      "mean_token_accuracy": 0.8872841037809849,
      "num_tokens": 9861471.0,
      "step": 100,
      "teacher_ce_loss": 2.50916
    },
    {
      "alignment_loss": 0.849671,
      "epoch": 0.352,
      "grad_norm": 419.76153564453125,
      "learning_rate": 8.392857142857144e-06,
      "loss": 42.0742,
      "mean_token_accuracy": 0.8809192053973675,
      "num_tokens": 10827332.0,
      "step": 110,
      "teacher_ce_loss": 2.530303
    },
    {
      "alignment_loss": 0.839253,
      "epoch": 0.384,
      "grad_norm": 403.6441345214844,
      "learning_rate": 8.23051948051948e-06,
      "loss": 42.7459,
      "mean_token_accuracy": 0.8831386599689722,
      "num_tokens": 11814144.0,
      "step": 120,
      "teacher_ce_loss": 2.503242
    },
    {
      "alignment_loss": 0.832565,
      "epoch": 0.416,
      "grad_norm": 458.2369384765625,
      "learning_rate": 8.068181818181819e-06,
      "loss": 43.0414,
      "mean_token_accuracy": 0.8847327880561352,
      "num_tokens": 12808535.0,
      "step": 130,
      "teacher_ce_loss": 2.526565
    },
    {
      "alignment_loss": 0.820745,
      "epoch": 0.448,
      "grad_norm": 6519.52978515625,
      "learning_rate": 7.905844155844157e-06,
      "loss": 40.8972,
      "mean_token_accuracy": 0.8860506009310484,
      "num_tokens": 13803034.0,
      "step": 140,
      "teacher_ce_loss": 2.56256
    },
    {
      "alignment_loss": 0.811062,
      "epoch": 0.48,
      "grad_norm": 411.56884765625,
      "learning_rate": 7.743506493506494e-06,
      "loss": 43.3087,
      "mean_token_accuracy": 0.8850693836808204,
      "num_tokens": 14792532.0,
      "step": 150,
      "teacher_ce_loss": 3.044439
    },
    {
      "alignment_loss": 0.801598,
      "epoch": 0.512,
      "grad_norm": 400.7724304199219,
      "learning_rate": 7.581168831168831e-06,
      "loss": 40.9337,
      "mean_token_accuracy": 0.8875109016895294,
      "num_tokens": 15762961.0,
      "step": 160,
      "teacher_ce_loss": 2.304987
    },
    {
      "alignment_loss": 0.791421,
      "epoch": 0.544,
      "grad_norm": 413.0471496582031,
      "learning_rate": 7.41883116883117e-06,
      "loss": 40.8531,
      "mean_token_accuracy": 0.887184364721179,
      "num_tokens": 16744298.0,
      "step": 170,
      "teacher_ce_loss": 2.488554
    },
    {
      "alignment_loss": 0.789574,
      "epoch": 0.576,
      "grad_norm": 416.1972351074219,
      "learning_rate": 7.2564935064935075e-06,
      "loss": 39.661,
      "mean_token_accuracy": 0.8898494675755501,
      "num_tokens": 17711917.0,
      "step": 180,
      "teacher_ce_loss": 2.359948
    },
    {
      "alignment_loss": 0.777883,
      "epoch": 0.608,
      "grad_norm": 546.523193359375,
      "learning_rate": 7.094155844155845e-06,
      "loss": 41.7644,
      "mean_token_accuracy": 0.885301411151886,
      "num_tokens": 18708337.0,
      "step": 190,
      "teacher_ce_loss": 3.000633
    },
    {
      "alignment_loss": 0.762139,
      "epoch": 0.64,
      "grad_norm": 408.8435363769531,
      "learning_rate": 6.931818181818183e-06,
      "loss": 38.7193,
      "mean_token_accuracy": 0.8898769337683916,
      "num_tokens": 19692844.0,
      "step": 200,
      "teacher_ce_loss": 2.393609
    }
  ],
  "logging_steps": 10,
  "max_steps": 626,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.152159340193382e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}