{
  "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": 0.9968553459119497,
      "grad_norm": 0.9062672853469849,
      "learning_rate": 1.8006289308176103e-05,
      "loss": 0.8042,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6709677419354839,
      "eval_loss": 0.4064599573612213,
      "eval_runtime": 1.4275,
      "eval_samples_per_second": 2171.572,
      "eval_steps_per_second": 45.533,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.7205497026443481,
      "learning_rate": 1.6012578616352204e-05,
      "loss": 0.3038,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.847741935483871,
      "eval_loss": 0.1362968385219574,
      "eval_runtime": 1.6105,
      "eval_samples_per_second": 1924.868,
      "eval_steps_per_second": 40.36,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.5987477898597717,
      "learning_rate": 1.4018867924528304e-05,
      "loss": 0.1395,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8990322580645161,
      "eval_loss": 0.07024983316659927,
      "eval_runtime": 1.6105,
      "eval_samples_per_second": 1924.879,
      "eval_steps_per_second": 40.36,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.5540674924850464,
      "learning_rate": 1.2025157232704403e-05,
      "loss": 0.0891,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9187096774193548,
      "eval_loss": 0.04933710768818855,
      "eval_runtime": 1.7991,
      "eval_samples_per_second": 1723.062,
      "eval_steps_per_second": 36.129,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.42864474654197693,
      "learning_rate": 1.0031446540880504e-05,
      "loss": 0.0692,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9241935483870968,
      "eval_loss": 0.04158218950033188,
      "eval_runtime": 1.6087,
      "eval_samples_per_second": 1927.079,
      "eval_steps_per_second": 40.406,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.4545074999332428,
      "learning_rate": 8.037735849056606e-06,
      "loss": 0.0595,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9270967741935484,
      "eval_loss": 0.03682653605937958,
      "eval_runtime": 1.4287,
      "eval_samples_per_second": 2169.74,
      "eval_steps_per_second": 45.495,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.34796932339668274,
      "learning_rate": 6.044025157232704e-06,
      "loss": 0.0538,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9316129032258065,
      "eval_loss": 0.03404370695352554,
      "eval_runtime": 1.6077,
      "eval_samples_per_second": 1928.227,
      "eval_steps_per_second": 40.431,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.3150351941585541,
      "learning_rate": 4.0503144654088055e-06,
      "loss": 0.0503,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9341935483870968,
      "eval_loss": 0.03234480321407318,
      "eval_runtime": 1.4217,
      "eval_samples_per_second": 2180.525,
      "eval_steps_per_second": 45.721,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.3166097104549408,
      "learning_rate": 2.056603773584906e-06,
      "loss": 0.0479,
      "step": 2853
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9341935483870968,
      "eval_loss": 0.03134315088391304,
      "eval_runtime": 1.6077,
      "eval_samples_per_second": 1928.261,
      "eval_steps_per_second": 40.431,
      "step": 2862
    },
    {
      "epoch": 9.968553459119496,
      "grad_norm": 0.27577438950538635,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.0467,
      "step": 3170
    }
  ],
  "logging_steps": 317,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 827333546055996.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.5781383032678951,
    "num_train_epochs": 10,
    "temperature": 2
  }
}