{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6270,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23923444976076555,
      "grad_norm": 1.7421875,
      "learning_rate": 1.8803827751196172e-05,
      "loss": 0.3483,
      "mean_token_accuracy": 0.9018868366479874,
      "step": 500
    },
    {
      "epoch": 0.4784688995215311,
      "grad_norm": 1.40625,
      "learning_rate": 1.7607655502392345e-05,
      "loss": 0.2415,
      "mean_token_accuracy": 0.9288070352077484,
      "step": 1000
    },
    {
      "epoch": 0.7177033492822966,
      "grad_norm": 1.5,
      "learning_rate": 1.641148325358852e-05,
      "loss": 0.2362,
      "mean_token_accuracy": 0.9295575091838837,
      "step": 1500
    },
    {
      "epoch": 0.9569377990430622,
      "grad_norm": 1.265625,
      "learning_rate": 1.5215311004784689e-05,
      "loss": 0.2303,
      "mean_token_accuracy": 0.9311141439676285,
      "step": 2000
    },
    {
      "epoch": 1.1961722488038278,
      "grad_norm": 1.109375,
      "learning_rate": 1.4019138755980863e-05,
      "loss": 0.2182,
      "mean_token_accuracy": 0.9329868614673614,
      "step": 2500
    },
    {
      "epoch": 1.4354066985645932,
      "grad_norm": 1.21875,
      "learning_rate": 1.2822966507177035e-05,
      "loss": 0.2112,
      "mean_token_accuracy": 0.9352543570995331,
      "step": 3000
    },
    {
      "epoch": 1.674641148325359,
      "grad_norm": 1.1640625,
      "learning_rate": 1.1626794258373206e-05,
      "loss": 0.217,
      "mean_token_accuracy": 0.9333346381187438,
      "step": 3500
    },
    {
      "epoch": 1.9138755980861244,
      "grad_norm": 1.5234375,
      "learning_rate": 1.0430622009569378e-05,
      "loss": 0.2096,
      "mean_token_accuracy": 0.9354861245155335,
      "step": 4000
    },
    {
      "epoch": 2.15311004784689,
      "grad_norm": 1.59375,
      "learning_rate": 9.234449760765552e-06,
      "loss": 0.2135,
      "mean_token_accuracy": 0.934171884059906,
      "step": 4500
    },
    {
      "epoch": 2.3923444976076556,
      "grad_norm": 1.0859375,
      "learning_rate": 8.038277511961722e-06,
      "loss": 0.2017,
      "mean_token_accuracy": 0.9373141669034958,
      "step": 5000
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 1.328125,
      "learning_rate": 6.842105263157896e-06,
      "loss": 0.2004,
      "mean_token_accuracy": 0.9374689855575562,
      "step": 5500
    },
    {
      "epoch": 2.8708133971291865,
      "grad_norm": 1.359375,
      "learning_rate": 5.645933014354067e-06,
      "loss": 0.2025,
      "mean_token_accuracy": 0.9373050066232681,
      "step": 6000
    }
  ],
  "logging_steps": 500,
  "max_steps": 8360,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.047045598855987e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}