{
  "best_metric": 0.9785056294779939,
  "best_model_checkpoint": "./results/checkpoint-32",
  "epoch": 8.0,
  "eval_steps": 16,
  "global_step": 32,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0,
      "eval_accuracy": 0.5972529906956137,
      "eval_f1": 0.26037428803905616,
      "eval_loss": 0.7424784898757935,
      "eval_precision": 0.6374501992031872,
      "eval_recall": 0.16359918200409,
      "eval_roc_auc": 0.5462249232928973,
      "eval_runtime": 102.4283,
      "eval_samples_per_second": 22.035,
      "eval_steps_per_second": 0.351,
      "step": 0
    },
    {
      "epoch": 0.25,
      "grad_norm": 494253.4375,
      "learning_rate": 5e-06,
      "loss": 0.6566,
      "step": 1
    },
    {
      "epoch": 0.5,
      "grad_norm": 401098.4375,
      "learning_rate": 1e-05,
      "loss": 0.691,
      "step": 2
    },
    {
      "epoch": 0.75,
      "grad_norm": 300705.59375,
      "learning_rate": 1.5e-05,
      "loss": 0.6193,
      "step": 3
    },
    {
      "epoch": 1.0,
      "grad_norm": 282439.5625,
      "learning_rate": 2e-05,
      "loss": 0.5831,
      "step": 4
    },
    {
      "epoch": 1.25,
      "grad_norm": 312752.5625,
      "learning_rate": 2.5e-05,
      "loss": 0.6175,
      "step": 5
    },
    {
      "epoch": 1.5,
      "grad_norm": 235615.21875,
      "learning_rate": 3e-05,
      "loss": 0.45,
      "step": 6
    },
    {
      "epoch": 1.75,
      "grad_norm": 319933.5,
      "learning_rate": 3.5e-05,
      "loss": 0.3565,
      "step": 7
    },
    {
      "epoch": 2.0,
      "grad_norm": 296006.78125,
      "learning_rate": 4e-05,
      "loss": 0.3097,
      "step": 8
    },
    {
      "epoch": 2.25,
      "grad_norm": 736140.375,
      "learning_rate": 4.5e-05,
      "loss": 0.4737,
      "step": 9
    },
    {
      "epoch": 2.5,
      "grad_norm": 228276.40625,
      "learning_rate": 5e-05,
      "loss": 0.1787,
      "step": 10
    },
    {
      "epoch": 2.75,
      "grad_norm": 346536.21875,
      "learning_rate": 4.772727272727273e-05,
      "loss": 0.1988,
      "step": 11
    },
    {
      "epoch": 3.0,
      "grad_norm": 467928.8125,
      "learning_rate": 4.545454545454546e-05,
      "loss": 0.2059,
      "step": 12
    },
    {
      "epoch": 3.25,
      "grad_norm": 198952.125,
      "learning_rate": 4.318181818181819e-05,
      "loss": 0.1082,
      "step": 13
    },
    {
      "epoch": 3.5,
      "grad_norm": 427120.96875,
      "learning_rate": 4.0909090909090915e-05,
      "loss": 0.1321,
      "step": 14
    },
    {
      "epoch": 3.75,
      "grad_norm": 586828.875,
      "learning_rate": 3.8636363636363636e-05,
      "loss": 0.2194,
      "step": 15
    },
    {
      "epoch": 4.0,
      "grad_norm": 316196.65625,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.1268,
      "step": 16
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8759415152857776,
      "eval_f1": 0.8360655737704918,
      "eval_loss": 0.26443684101104736,
      "eval_precision": 0.9780821917808219,
      "eval_recall": 0.7300613496932515,
      "eval_roc_auc": 0.858775788216446,
      "eval_runtime": 102.1683,
      "eval_samples_per_second": 22.091,
      "eval_steps_per_second": 0.352,
      "step": 16
    },
    {
      "epoch": 4.25,
      "grad_norm": 290404.625,
      "learning_rate": 3.409090909090909e-05,
      "loss": 0.0596,
      "step": 17
    },
    {
      "epoch": 4.5,
      "grad_norm": 183575.28125,
      "learning_rate": 3.181818181818182e-05,
      "loss": 0.0991,
      "step": 18
    },
    {
      "epoch": 4.75,
      "grad_norm": 52958.6171875,
      "learning_rate": 2.954545454545455e-05,
      "loss": 0.0233,
      "step": 19
    },
    {
      "epoch": 5.0,
      "grad_norm": 366309.96875,
      "learning_rate": 2.7272727272727273e-05,
      "loss": 0.0678,
      "step": 20
    },
    {
      "epoch": 5.25,
      "grad_norm": 133315.859375,
      "learning_rate": 2.5e-05,
      "loss": 0.0393,
      "step": 21
    },
    {
      "epoch": 5.5,
      "grad_norm": 63111.57421875,
      "learning_rate": 2.272727272727273e-05,
      "loss": 0.0136,
      "step": 22
    },
    {
      "epoch": 5.75,
      "grad_norm": 58023.6875,
      "learning_rate": 2.0454545454545457e-05,
      "loss": 0.0117,
      "step": 23
    },
    {
      "epoch": 6.0,
      "grad_norm": 58158.015625,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.0143,
      "step": 24
    },
    {
      "epoch": 6.25,
      "grad_norm": 27386.134765625,
      "learning_rate": 1.590909090909091e-05,
      "loss": 0.0077,
      "step": 25
    },
    {
      "epoch": 6.5,
      "grad_norm": 38601.58203125,
      "learning_rate": 1.3636363636363637e-05,
      "loss": 0.0084,
      "step": 26
    },
    {
      "epoch": 6.75,
      "grad_norm": 241114.9375,
      "learning_rate": 1.1363636363636365e-05,
      "loss": 0.2064,
      "step": 27
    },
    {
      "epoch": 7.0,
      "grad_norm": 53962.51953125,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.012,
      "step": 28
    },
    {
      "epoch": 7.25,
      "grad_norm": 20690.81640625,
      "learning_rate": 6.818181818181818e-06,
      "loss": 0.0069,
      "step": 29
    },
    {
      "epoch": 7.5,
      "grad_norm": 8544.26171875,
      "learning_rate": 4.5454545454545455e-06,
      "loss": 0.0036,
      "step": 30
    },
    {
      "epoch": 7.75,
      "grad_norm": 5743.671875,
      "learning_rate": 2.2727272727272728e-06,
      "loss": 0.0029,
      "step": 31
    },
    {
      "epoch": 8.0,
      "grad_norm": 85053.53125,
      "learning_rate": 0.0,
      "loss": 0.009,
      "step": 32
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9813912272928667,
      "eval_f1": 0.9785056294779939,
      "eval_loss": 0.04325682297348976,
      "eval_precision": 0.9795081967213115,
      "eval_recall": 0.9775051124744376,
      "eval_roc_auc": 0.9809339479494941,
      "eval_runtime": 102.2325,
      "eval_samples_per_second": 22.077,
      "eval_steps_per_second": 0.352,
      "step": 32
    }
  ],
  "logging_steps": 1,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 16,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.507152629825536e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}