{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.433962264150944,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6770967741935484,
      "eval_loss": 0.4036009609699249,
      "eval_runtime": 5.5524,
      "eval_samples_per_second": 558.319,
      "eval_steps_per_second": 11.707,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.8646998405456543,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.6406,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8458064516129032,
      "eval_loss": 0.13483001291751862,
      "eval_runtime": 5.6164,
      "eval_samples_per_second": 551.959,
      "eval_steps_per_second": 11.573,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8977419354838709,
      "eval_loss": 0.0700574740767479,
      "eval_runtime": 5.6096,
      "eval_samples_per_second": 552.623,
      "eval_steps_per_second": 11.587,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.6868194937705994,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1593,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9206451612903226,
      "eval_loss": 0.049840863794088364,
      "eval_runtime": 5.565,
      "eval_samples_per_second": 557.05,
      "eval_steps_per_second": 11.68,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.4269469082355499,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0792,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9261290322580645,
      "eval_loss": 0.0415181890130043,
      "eval_runtime": 5.7105,
      "eval_samples_per_second": 542.864,
      "eval_steps_per_second": 11.383,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9293548387096774,
      "eval_loss": 0.03712046518921852,
      "eval_runtime": 5.26,
      "eval_samples_per_second": 589.35,
      "eval_steps_per_second": 12.357,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.32833540439605713,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0593,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9348387096774193,
      "eval_loss": 0.03438512980937958,
      "eval_runtime": 5.6841,
      "eval_samples_per_second": 545.377,
      "eval_steps_per_second": 11.435,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.37838014960289,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0514,
      "step": 2500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9319354838709677,
      "eval_loss": 0.03265618532896042,
      "eval_runtime": 5.2992,
      "eval_samples_per_second": 584.991,
      "eval_steps_per_second": 12.266,
      "step": 2544
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9335483870967742,
      "eval_loss": 0.03140486776828766,
      "eval_runtime": 5.3205,
      "eval_samples_per_second": 582.655,
      "eval_steps_per_second": 12.217,
      "step": 2862
    },
    {
      "epoch": 9.433962264150944,
      "grad_norm": 0.31082895398139954,
      "learning_rate": 1.1320754716981133e-06,
      "loss": 0.0478,
      "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 875211304907184.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6190430442458457,
    "num_train_epochs": 10,
    "temperature": 2
  }
}