{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2862,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9968553459119497,
      "grad_norm": 0.965503990650177,
      "learning_rate": 1.7784765897973445e-05,
      "loss": 0.8047,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6812903225806451,
      "eval_loss": 0.4058631658554077,
      "eval_runtime": 1.6943,
      "eval_samples_per_second": 1829.632,
      "eval_steps_per_second": 38.363,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.7085171341896057,
      "learning_rate": 1.556953179594689e-05,
      "loss": 0.3062,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8503225806451613,
      "eval_loss": 0.13832151889801025,
      "eval_runtime": 1.7583,
      "eval_samples_per_second": 1763.081,
      "eval_steps_per_second": 36.968,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.619353711605072,
      "learning_rate": 1.3354297693920338e-05,
      "loss": 0.1421,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8958064516129032,
      "eval_loss": 0.07200490683317184,
      "eval_runtime": 1.7884,
      "eval_samples_per_second": 1733.388,
      "eval_steps_per_second": 36.345,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.5501407384872437,
      "learning_rate": 1.1139063591893781e-05,
      "loss": 0.0916,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9125806451612903,
      "eval_loss": 0.05142820626497269,
      "eval_runtime": 1.7858,
      "eval_samples_per_second": 1735.903,
      "eval_steps_per_second": 36.398,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.3894004821777344,
      "learning_rate": 8.923829489867226e-06,
      "loss": 0.0717,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.925483870967742,
      "eval_loss": 0.04311361908912659,
      "eval_runtime": 1.7985,
      "eval_samples_per_second": 1723.616,
      "eval_steps_per_second": 36.14,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.4861418902873993,
      "learning_rate": 6.708595387840672e-06,
      "loss": 0.0615,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9261290322580645,
      "eval_loss": 0.03821605071425438,
      "eval_runtime": 1.8068,
      "eval_samples_per_second": 1715.761,
      "eval_steps_per_second": 35.976,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.3920305669307709,
      "learning_rate": 4.4933612858141165e-06,
      "loss": 0.0561,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9325806451612904,
      "eval_loss": 0.035470735281705856,
      "eval_runtime": 1.7961,
      "eval_samples_per_second": 1725.979,
      "eval_steps_per_second": 36.19,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.3401239514350891,
      "learning_rate": 2.2781271837875614e-06,
      "loss": 0.0528,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9316129032258065,
      "eval_loss": 0.03403536230325699,
      "eval_runtime": 1.8035,
      "eval_samples_per_second": 1718.877,
      "eval_steps_per_second": 36.041,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.3334876596927643,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.0511,
      "step": 2853
    }
  ],
  "logging_steps": 317,
  "max_steps": 2862,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 1000000000.0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 742838622388212.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6623158454366724,
    "num_train_epochs": 9,
    "temperature": 2
  }
}