{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.861635220125786,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5951612903225807,
      "eval_loss": 0.20783506333827972,
      "eval_runtime": 1.3226,
      "eval_samples_per_second": 2343.937,
      "eval_steps_per_second": 49.147,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5312509536743164,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.3341,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8338709677419355,
      "eval_loss": 0.09971419721841812,
      "eval_runtime": 1.3325,
      "eval_samples_per_second": 2326.534,
      "eval_steps_per_second": 48.782,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.882258064516129,
      "eval_loss": 0.06562549620866776,
      "eval_runtime": 1.3417,
      "eval_samples_per_second": 2310.56,
      "eval_steps_per_second": 48.447,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.4943838119506836,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1156,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8958064516129032,
      "eval_loss": 0.04985633119940758,
      "eval_runtime": 1.3449,
      "eval_samples_per_second": 2305.032,
      "eval_steps_per_second": 48.331,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.3221791386604309,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0725,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9096774193548387,
      "eval_loss": 0.04061922803521156,
      "eval_runtime": 1.3564,
      "eval_samples_per_second": 2285.502,
      "eval_steps_per_second": 47.922,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9187096774193548,
      "eval_loss": 0.03560462221503258,
      "eval_runtime": 1.3515,
      "eval_samples_per_second": 2293.716,
      "eval_steps_per_second": 48.094,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.2550632953643799,
      "learning_rate": 6.02375960866527e-06,
      "loss": 0.0569,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.03291216120123863,
      "eval_runtime": 1.3475,
      "eval_samples_per_second": 2300.507,
      "eval_steps_per_second": 48.236,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.2292759269475937,
      "learning_rate": 2.5296995108315863e-06,
      "loss": 0.0497,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7941624119034844,
    "num_train_epochs": 9,
    "temperature": 8
  }
}