{
  "best_metric": 0.7926666666666666,
  "best_model_checkpoint": "Bert_tinybert-distilled/run-2/checkpoint-1720",
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 1720,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.39246517419815063,
      "learning_rate": 4.639522353304605e-05,
      "loss": 0.2489,
      "step": 172
    },
    {
      "epoch": 1.0,
      "eval_AUC": 0.786481777777778,
      "eval_BAC": 0.6966666666666667,
      "eval_MCC": 0.4004741770824115,
      "eval_PR_AUC": 0.7486102682312443,
      "eval_accuracy": 0.6966666666666667,
      "eval_loss": 0.23812349140644073,
      "eval_runtime": 6.2157,
      "eval_samples_per_second": 241.324,
      "eval_steps_per_second": 7.561,
      "step": 172
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.45007961988449097,
      "learning_rate": 4.1240198696040935e-05,
      "loss": 0.2402,
      "step": 344
    },
    {
      "epoch": 2.0,
      "eval_AUC": 0.8126222222222222,
      "eval_BAC": 0.7553333333333334,
      "eval_MCC": 0.5408380051918336,
      "eval_PR_AUC": 0.7628286445519327,
      "eval_accuracy": 0.7553333333333333,
      "eval_loss": 0.23252661526203156,
      "eval_runtime": 6.2733,
      "eval_samples_per_second": 239.107,
      "eval_steps_per_second": 7.492,
      "step": 344
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.5242908596992493,
      "learning_rate": 3.608517385903581e-05,
      "loss": 0.2384,
      "step": 516
    },
    {
      "epoch": 3.0,
      "eval_AUC": 0.8210719999999999,
      "eval_BAC": 0.754,
      "eval_MCC": 0.5442085309980096,
      "eval_PR_AUC": 0.7771659186047205,
      "eval_accuracy": 0.754,
      "eval_loss": 0.23356583714485168,
      "eval_runtime": 6.2379,
      "eval_samples_per_second": 240.465,
      "eval_steps_per_second": 7.535,
      "step": 516
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7818685173988342,
      "learning_rate": 3.0930149022030695e-05,
      "loss": 0.237,
      "step": 688
    },
    {
      "epoch": 4.0,
      "eval_AUC": 0.8297351111111111,
      "eval_BAC": 0.7686666666666666,
      "eval_MCC": 0.5525864831739952,
      "eval_PR_AUC": 0.7854176937233266,
      "eval_accuracy": 0.7686666666666667,
      "eval_loss": 0.2301863133907318,
      "eval_runtime": 6.2464,
      "eval_samples_per_second": 240.136,
      "eval_steps_per_second": 7.524,
      "step": 688
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.7255853414535522,
      "learning_rate": 2.577512418502558e-05,
      "loss": 0.2362,
      "step": 860
    },
    {
      "epoch": 5.0,
      "eval_AUC": 0.836256,
      "eval_BAC": 0.7666666666666666,
      "eval_MCC": 0.5542293013644021,
      "eval_PR_AUC": 0.7972297123276614,
      "eval_accuracy": 0.7666666666666667,
      "eval_loss": 0.22993715107440948,
      "eval_runtime": 6.2748,
      "eval_samples_per_second": 239.051,
      "eval_steps_per_second": 7.49,
      "step": 860
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.7448613047599792,
      "learning_rate": 2.0620099348020468e-05,
      "loss": 0.235,
      "step": 1032
    },
    {
      "epoch": 6.0,
      "eval_AUC": 0.8438133333333333,
      "eval_BAC": 0.778,
      "eval_MCC": 0.5714092913131209,
      "eval_PR_AUC": 0.8061543385382132,
      "eval_accuracy": 0.778,
      "eval_loss": 0.22893689572811127,
      "eval_runtime": 6.2414,
      "eval_samples_per_second": 240.331,
      "eval_steps_per_second": 7.53,
      "step": 1032
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7925253510475159,
      "learning_rate": 1.5465074511015347e-05,
      "loss": 0.2345,
      "step": 1204
    },
    {
      "epoch": 7.0,
      "eval_AUC": 0.8473991111111112,
      "eval_BAC": 0.782,
      "eval_MCC": 0.5792569729386494,
      "eval_PR_AUC": 0.8108360182006333,
      "eval_accuracy": 0.782,
      "eval_loss": 0.2287604957818985,
      "eval_runtime": 6.2713,
      "eval_samples_per_second": 239.186,
      "eval_steps_per_second": 7.494,
      "step": 1204
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.35333141684532166,
      "learning_rate": 1.0310049674010234e-05,
      "loss": 0.234,
      "step": 1376
    },
    {
      "epoch": 8.0,
      "eval_AUC": 0.8490613333333333,
      "eval_BAC": 0.788,
      "eval_MCC": 0.5830155950947427,
      "eval_PR_AUC": 0.8120541754208705,
      "eval_accuracy": 0.788,
      "eval_loss": 0.2279704362154007,
      "eval_runtime": 6.2416,
      "eval_samples_per_second": 240.322,
      "eval_steps_per_second": 7.53,
      "step": 1376
    },
    {
      "epoch": 9.0,
      "grad_norm": 0.44008931517601013,
      "learning_rate": 5.155024837005117e-06,
      "loss": 0.2331,
      "step": 1548
    },
    {
      "epoch": 9.0,
      "eval_AUC": 0.8531626666666667,
      "eval_BAC": 0.79,
      "eval_MCC": 0.5917962039710227,
      "eval_PR_AUC": 0.81588940138238,
      "eval_accuracy": 0.79,
      "eval_loss": 0.22790811955928802,
      "eval_runtime": 6.2673,
      "eval_samples_per_second": 239.338,
      "eval_steps_per_second": 7.499,
      "step": 1548
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.5285153388977051,
      "learning_rate": 0.0,
      "loss": 0.2334,
      "step": 1720
    },
    {
      "epoch": 10.0,
      "eval_AUC": 0.8541422222222221,
      "eval_BAC": 0.7926666666666666,
      "eval_MCC": 0.5969110830375018,
      "eval_PR_AUC": 0.8163240976491174,
      "eval_accuracy": 0.7926666666666666,
      "eval_loss": 0.22778959572315216,
      "eval_runtime": 6.2536,
      "eval_samples_per_second": 239.861,
      "eval_steps_per_second": 7.516,
      "step": 1720
    }
  ],
  "logging_steps": 500,
  "max_steps": 1720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4503777300000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.3558486326039224,
    "learning_rate": 5.155024837005116e-05,
    "num_train_epochs": 10,
    "temperature": 4
  }
}