{
  "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": 0.6502413153648376,
      "learning_rate": 1.8006289308176103e-05,
      "loss": 0.4238,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6119354838709677,
      "eval_loss": 0.20929040014743805,
      "eval_runtime": 2.5206,
      "eval_samples_per_second": 1229.853,
      "eval_steps_per_second": 25.787,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.5239719152450562,
      "learning_rate": 1.6012578616352204e-05,
      "loss": 0.1738,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8325806451612904,
      "eval_loss": 0.09838643670082092,
      "eval_runtime": 2.6045,
      "eval_samples_per_second": 1190.262,
      "eval_steps_per_second": 24.957,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.5191203355789185,
      "learning_rate": 1.4018867924528304e-05,
      "loss": 0.106,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.885483870967742,
      "eval_loss": 0.06351756304502487,
      "eval_runtime": 2.5551,
      "eval_samples_per_second": 1213.263,
      "eval_steps_per_second": 25.439,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.3775458037853241,
      "learning_rate": 1.2025157232704403e-05,
      "loss": 0.0776,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.04740825667977333,
      "eval_runtime": 2.6242,
      "eval_samples_per_second": 1181.332,
      "eval_steps_per_second": 24.77,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.28506648540496826,
      "learning_rate": 1.0031446540880504e-05,
      "loss": 0.0633,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9116129032258065,
      "eval_loss": 0.03853140398859978,
      "eval_runtime": 2.7766,
      "eval_samples_per_second": 1116.461,
      "eval_steps_per_second": 23.41,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.30514639616012573,
      "learning_rate": 8.037735849056606e-06,
      "loss": 0.0548,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9190322580645162,
      "eval_loss": 0.03350428119301796,
      "eval_runtime": 2.7999,
      "eval_samples_per_second": 1107.165,
      "eval_steps_per_second": 23.215,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.3329088091850281,
      "learning_rate": 6.044025157232704e-06,
      "loss": 0.0494,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9251612903225807,
      "eval_loss": 0.030477993190288544,
      "eval_runtime": 2.7684,
      "eval_samples_per_second": 1119.781,
      "eval_steps_per_second": 23.479,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.29392293095588684,
      "learning_rate": 4.0503144654088055e-06,
      "loss": 0.046,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9258064516129032,
      "eval_loss": 0.028436731547117233,
      "eval_runtime": 2.7553,
      "eval_samples_per_second": 1125.094,
      "eval_steps_per_second": 23.591,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.4509568512439728,
      "learning_rate": 2.056603773584906e-06,
      "loss": 0.0439,
      "step": 2853
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.93,
      "eval_loss": 0.02739552967250347,
      "eval_runtime": 2.6928,
      "eval_samples_per_second": 1151.209,
      "eval_steps_per_second": 24.138,
      "step": 2862
    },
    {
      "epoch": 9.968553459119496,
      "grad_norm": 0.273268461227417,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.0429,
      "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": 825891210476556.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.16553214790840132,
    "num_train_epochs": 10,
    "temperature": 7
  }
}