{
  "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": 0.6643747091293335,
      "learning_rate": 1.7784765897973445e-05,
      "loss": 0.4592,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6206451612903225,
      "eval_loss": 0.22760848701000214,
      "eval_runtime": 3.3418,
      "eval_samples_per_second": 927.649,
      "eval_steps_per_second": 19.451,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.5484316349029541,
      "learning_rate": 1.556953179594689e-05,
      "loss": 0.1875,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.837741935483871,
      "eval_loss": 0.10404686629772186,
      "eval_runtime": 3.1167,
      "eval_samples_per_second": 994.653,
      "eval_steps_per_second": 20.856,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.5544776916503906,
      "learning_rate": 1.3354297693920338e-05,
      "loss": 0.112,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.885483870967742,
      "eval_loss": 0.06577753275632858,
      "eval_runtime": 3.2395,
      "eval_samples_per_second": 956.946,
      "eval_steps_per_second": 20.065,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.4056822657585144,
      "learning_rate": 1.1139063591893781e-05,
      "loss": 0.0813,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9012903225806451,
      "eval_loss": 0.04862780496478081,
      "eval_runtime": 3.8223,
      "eval_samples_per_second": 811.026,
      "eval_steps_per_second": 17.005,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.3019745349884033,
      "learning_rate": 8.923829489867226e-06,
      "loss": 0.0662,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9109677419354839,
      "eval_loss": 0.03964576870203018,
      "eval_runtime": 4.5329,
      "eval_samples_per_second": 683.893,
      "eval_steps_per_second": 14.34,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.3191016614437103,
      "learning_rate": 6.708595387840672e-06,
      "loss": 0.0575,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9187096774193548,
      "eval_loss": 0.0346621572971344,
      "eval_runtime": 3.0634,
      "eval_samples_per_second": 1011.932,
      "eval_steps_per_second": 21.218,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.3592301905155182,
      "learning_rate": 4.4933612858141165e-06,
      "loss": 0.0522,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9235483870967742,
      "eval_loss": 0.03190623223781586,
      "eval_runtime": 3.0567,
      "eval_samples_per_second": 1014.167,
      "eval_steps_per_second": 21.265,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.32463663816452026,
      "learning_rate": 2.2781271837875614e-06,
      "loss": 0.0491,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9251612903225807,
      "eval_loss": 0.0301646888256073,
      "eval_runtime": 3.0291,
      "eval_samples_per_second": 1023.403,
      "eval_steps_per_second": 21.458,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.5000515580177307,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.0476,
      "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": 745151547572796.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9906482796828688,
    "num_train_epochs": 9,
    "temperature": 5
  }
}