{
  "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": 0.5887096774193549,
      "eval_loss": 0.19886387884616852,
      "eval_runtime": 5.1732,
      "eval_samples_per_second": 599.241,
      "eval_steps_per_second": 12.565,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5294517874717712,
      "learning_rate": 1.650593990216632e-05,
      "loss": 0.3203,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8193548387096774,
      "eval_loss": 0.09618715196847916,
      "eval_runtime": 5.2244,
      "eval_samples_per_second": 593.365,
      "eval_steps_per_second": 12.442,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8783870967741936,
      "eval_loss": 0.06421981006860733,
      "eval_runtime": 5.1205,
      "eval_samples_per_second": 605.405,
      "eval_steps_per_second": 12.694,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.44660821557044983,
      "learning_rate": 1.3011879804332637e-05,
      "loss": 0.1116,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8996774193548387,
      "eval_loss": 0.04962550476193428,
      "eval_runtime": 5.0817,
      "eval_samples_per_second": 610.038,
      "eval_steps_per_second": 12.791,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.3033629357814789,
      "learning_rate": 9.517819706498952e-06,
      "loss": 0.0713,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9080645161290323,
      "eval_loss": 0.040737707167863846,
      "eval_runtime": 5.3814,
      "eval_samples_per_second": 576.062,
      "eval_steps_per_second": 12.079,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9116129032258065,
      "eval_loss": 0.0357489213347435,
      "eval_runtime": 5.0778,
      "eval_samples_per_second": 610.498,
      "eval_steps_per_second": 12.801,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.26608094573020935,
      "learning_rate": 6.02375960866527e-06,
      "loss": 0.0561,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9187096774193548,
      "eval_loss": 0.03307666257023811,
      "eval_runtime": 5.171,
      "eval_samples_per_second": 599.502,
      "eval_steps_per_second": 12.57,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.27350321412086487,
      "learning_rate": 2.5296995108315863e-06,
      "loss": 0.0492,
      "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": 760794662700744.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.260556849082234,
    "num_train_epochs": 9,
    "temperature": 11
  }
}