{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 3180,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5838709677419355,
      "eval_loss": 0.19346614181995392,
      "eval_runtime": 5.4448,
      "eval_samples_per_second": 569.354,
      "eval_steps_per_second": 11.938,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5179193019866943,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.3125,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.09356740862131119,
      "eval_runtime": 5.7422,
      "eval_samples_per_second": 539.859,
      "eval_steps_per_second": 11.32,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8790322580645161,
      "eval_loss": 0.06245172396302223,
      "eval_runtime": 5.6885,
      "eval_samples_per_second": 544.959,
      "eval_steps_per_second": 11.427,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.4341842532157898,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1085,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9006451612903226,
      "eval_loss": 0.048103515058755875,
      "eval_runtime": 5.5558,
      "eval_samples_per_second": 557.971,
      "eval_steps_per_second": 11.699,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.29487845301628113,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.069,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9093548387096774,
      "eval_loss": 0.03922608867287636,
      "eval_runtime": 5.6358,
      "eval_samples_per_second": 550.056,
      "eval_steps_per_second": 11.533,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.912258064516129,
      "eval_loss": 0.03417789563536644,
      "eval_runtime": 5.7261,
      "eval_samples_per_second": 541.383,
      "eval_steps_per_second": 11.352,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.2546312212944031,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0537,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9219354838709677,
      "eval_loss": 0.03144192695617676,
      "eval_runtime": 5.8726,
      "eval_samples_per_second": 527.873,
      "eval_steps_per_second": 11.068,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.2692946791648865,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0465,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9267741935483871,
      "eval_loss": 0.029110092669725418,
      "eval_runtime": 5.7312,
      "eval_samples_per_second": 540.899,
      "eval_steps_per_second": 11.341,
      "step": 2544
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9283870967741935,
      "eval_loss": 0.028060223907232285,
      "eval_runtime": 5.5233,
      "eval_samples_per_second": 561.26,
      "eval_steps_per_second": 11.768,
      "step": 2862
    },
    {
      "epoch": 9.433962264150944,
      "grad_norm": 0.2502853572368622,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0431,
      "step": 3000
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 780738843279612.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.022235028872426232,
    "num_train_epochs": 10,
    "temperature": 14
  }
}