{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.24647887323943662,
  "eval_steps": 500,
  "global_step": 35,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.035211267605633804,
      "grad_norm": 4.435381889343262,
      "learning_rate": 8.401166180758017e-06,
      "loss": 0.3742,
      "mean_token_accuracy": 0.8945995926856994,
      "num_tokens": 301695.0,
      "step": 5
    },
    {
      "epoch": 0.07042253521126761,
      "grad_norm": 1.0536936521530151,
      "learning_rate": 1.8902623906705538e-05,
      "loss": 0.2437,
      "mean_token_accuracy": 0.9109556913375855,
      "num_tokens": 607230.0,
      "step": 10
    },
    {
      "epoch": 0.1056338028169014,
      "grad_norm": 0.8927873373031616,
      "learning_rate": 2.9404081632653062e-05,
      "loss": 0.201,
      "mean_token_accuracy": 0.9295879244804383,
      "num_tokens": 913896.0,
      "step": 15
    },
    {
      "epoch": 0.14084507042253522,
      "grad_norm": 0.6603397130966187,
      "learning_rate": 3.990553935860058e-05,
      "loss": 0.1947,
      "mean_token_accuracy": 0.9296464204788208,
      "num_tokens": 1234009.0,
      "step": 20
    },
    {
      "epoch": 0.176056338028169,
      "grad_norm": 0.5713601112365723,
      "learning_rate": 5.0406997084548104e-05,
      "loss": 0.1712,
      "mean_token_accuracy": 0.9373371362686157,
      "num_tokens": 1536066.0,
      "step": 25
    },
    {
      "epoch": 0.2112676056338028,
      "grad_norm": 0.6312325596809387,
      "learning_rate": 6.090845481049563e-05,
      "loss": 0.1683,
      "mean_token_accuracy": 0.9364290118217469,
      "num_tokens": 1854154.0,
      "step": 30
    },
    {
      "epoch": 0.24647887323943662,
      "grad_norm": 1.2735894918441772,
      "learning_rate": 7.140991253644314e-05,
      "loss": 0.1312,
      "mean_token_accuracy": 0.9548299431800842,
      "num_tokens": 2160408.0,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 142,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.105089563623424e+16,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}