{
  "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": 0.6225002408027649,
      "learning_rate": 1.7784765897973445e-05,
      "loss": 0.3828,
      "step": 317
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5812903225806452,
      "eval_loss": 0.18950515985488892,
      "eval_runtime": 2.808,
      "eval_samples_per_second": 1103.991,
      "eval_steps_per_second": 23.148,
      "step": 318
    },
    {
      "epoch": 1.9937106918238994,
      "grad_norm": 0.5088759064674377,
      "learning_rate": 1.556953179594689e-05,
      "loss": 0.1594,
      "step": 634
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8183870967741935,
      "eval_loss": 0.09329316020011902,
      "eval_runtime": 3.1063,
      "eval_samples_per_second": 997.977,
      "eval_steps_per_second": 20.925,
      "step": 636
    },
    {
      "epoch": 2.990566037735849,
      "grad_norm": 0.4906318783760071,
      "learning_rate": 1.3354297693920338e-05,
      "loss": 0.1006,
      "step": 951
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8793548387096775,
      "eval_loss": 0.06283030658960342,
      "eval_runtime": 4.811,
      "eval_samples_per_second": 644.359,
      "eval_steps_per_second": 13.511,
      "step": 954
    },
    {
      "epoch": 3.9874213836477987,
      "grad_norm": 0.3708745539188385,
      "learning_rate": 1.1139063591893781e-05,
      "loss": 0.0757,
      "step": 1268
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8961290322580645,
      "eval_loss": 0.04845510795712471,
      "eval_runtime": 3.0191,
      "eval_samples_per_second": 1026.797,
      "eval_steps_per_second": 21.53,
      "step": 1272
    },
    {
      "epoch": 4.984276729559748,
      "grad_norm": 0.27722346782684326,
      "learning_rate": 8.923829489867226e-06,
      "loss": 0.0629,
      "step": 1585
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9064516129032258,
      "eval_loss": 0.040148768573999405,
      "eval_runtime": 3.1116,
      "eval_samples_per_second": 996.286,
      "eval_steps_per_second": 20.89,
      "step": 1590
    },
    {
      "epoch": 5.981132075471698,
      "grad_norm": 0.2924395799636841,
      "learning_rate": 6.708595387840672e-06,
      "loss": 0.0553,
      "step": 1902
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9119354838709678,
      "eval_loss": 0.035382989794015884,
      "eval_runtime": 3.3223,
      "eval_samples_per_second": 933.091,
      "eval_steps_per_second": 19.565,
      "step": 1908
    },
    {
      "epoch": 6.977987421383648,
      "grad_norm": 0.3113860785961151,
      "learning_rate": 4.4933612858141165e-06,
      "loss": 0.0504,
      "step": 2219
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.917741935483871,
      "eval_loss": 0.032723989337682724,
      "eval_runtime": 3.2168,
      "eval_samples_per_second": 963.691,
      "eval_steps_per_second": 20.206,
      "step": 2226
    },
    {
      "epoch": 7.9748427672955975,
      "grad_norm": 0.2782796621322632,
      "learning_rate": 2.2781271837875614e-06,
      "loss": 0.0475,
      "step": 2536
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9212903225806451,
      "eval_loss": 0.031014930456876755,
      "eval_runtime": 3.1818,
      "eval_samples_per_second": 974.289,
      "eval_steps_per_second": 20.429,
      "step": 2544
    },
    {
      "epoch": 8.971698113207546,
      "grad_norm": 0.4069852828979492,
      "learning_rate": 6.289308176100629e-08,
      "loss": 0.0461,
      "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": 0.7922936555306777,
    "num_train_epochs": 9,
    "temperature": 18
  }
}