{
  "best_metric": 0.8173333333333334,
  "best_model_checkpoint": "Bert_tinybert-distilled/run-31/checkpoint-688",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1032,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9708962440490723,
      "learning_rate": 0.00018007353154788933,
      "loss": 0.5186,
      "step": 172
    },
    {
      "epoch": 1.0,
      "eval_AUC": 0.8338702222222223,
      "eval_BAC": 0.758,
      "eval_MCC": 0.5164413442563544,
      "eval_PR_AUC": 0.7903470719520844,
      "eval_accuracy": 0.758,
      "eval_loss": 0.45808151364326477,
      "eval_runtime": 6.2856,
      "eval_samples_per_second": 238.642,
      "eval_steps_per_second": 7.477,
      "step": 172
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5569114685058594,
      "learning_rate": 0.0001543487413267623,
      "loss": 0.4771,
      "step": 344
    },
    {
      "epoch": 2.0,
      "eval_AUC": 0.8571271111111111,
      "eval_BAC": 0.7793333333333333,
      "eval_MCC": 0.591094330111451,
      "eval_PR_AUC": 0.8037547287750328,
      "eval_accuracy": 0.7793333333333333,
      "eval_loss": 0.4492648243904114,
      "eval_runtime": 6.2364,
      "eval_samples_per_second": 240.525,
      "eval_steps_per_second": 7.536,
      "step": 344
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.398871898651123,
      "learning_rate": 0.00012862395110563524,
      "loss": 0.4628,
      "step": 516
    },
    {
      "epoch": 3.0,
      "eval_AUC": 0.8642417777777778,
      "eval_BAC": 0.7986666666666666,
      "eval_MCC": 0.5973864367101849,
      "eval_PR_AUC": 0.8284732954100074,
      "eval_accuracy": 0.7986666666666666,
      "eval_loss": 0.425106942653656,
      "eval_runtime": 6.229,
      "eval_samples_per_second": 240.809,
      "eval_steps_per_second": 7.545,
      "step": 516
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.736416816711426,
      "learning_rate": 0.0001028991608845082,
      "loss": 0.453,
      "step": 688
    },
    {
      "epoch": 4.0,
      "eval_AUC": 0.8754026666666667,
      "eval_BAC": 0.8173333333333332,
      "eval_MCC": 0.6381270399870552,
      "eval_PR_AUC": 0.8391884252076478,
      "eval_accuracy": 0.8173333333333334,
      "eval_loss": 0.4091016352176666,
      "eval_runtime": 6.2348,
      "eval_samples_per_second": 240.585,
      "eval_steps_per_second": 7.538,
      "step": 688
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.7166924476623535,
      "learning_rate": 7.717437066338115e-05,
      "loss": 0.4436,
      "step": 860
    },
    {
      "epoch": 5.0,
      "eval_AUC": 0.8772906666666666,
      "eval_BAC": 0.8113333333333334,
      "eval_MCC": 0.6330057971036138,
      "eval_PR_AUC": 0.8365753409728844,
      "eval_accuracy": 0.8113333333333334,
      "eval_loss": 0.4114744961261749,
      "eval_runtime": 6.2226,
      "eval_samples_per_second": 241.056,
      "eval_steps_per_second": 7.553,
      "step": 860
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.523271560668945,
      "learning_rate": 5.14495804422541e-05,
      "loss": 0.4382,
      "step": 1032
    },
    {
      "epoch": 6.0,
      "eval_AUC": 0.8797031111111111,
      "eval_BAC": 0.8173333333333334,
      "eval_MCC": 0.6408579236799679,
      "eval_PR_AUC": 0.8454815799054289,
      "eval_accuracy": 0.8173333333333334,
      "eval_loss": 0.4119183421134949,
      "eval_runtime": 6.233,
      "eval_samples_per_second": 240.656,
      "eval_steps_per_second": 7.541,
      "step": 1032
    }
  ],
  "logging_steps": 500,
  "max_steps": 1376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2702266380000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8022374115576867,
    "learning_rate": 0.0002057983217690164,
    "num_train_epochs": 8,
    "temperature": 4
  }
}