{
  "best_metric": 0.8941450174486235,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-5160",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 5160,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9960685968399048,
      "learning_rate": 0.0005204222579442675,
      "loss": 0.3335,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8900736719658783,
      "eval_f1": 0.0070052539404553416,
      "eval_loss": 0.3329962193965912,
      "eval_mcc": 0.05593214040002362,
      "eval_precision": 1.0,
      "eval_recall": 0.0035149384885764497,
      "eval_runtime": 9.9674,
      "eval_samples_per_second": 517.487,
      "eval_steps_per_second": 16.253,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6790906190872192,
      "learning_rate": 0.0004625975626171267,
      "loss": 0.3255,
      "step": 1290
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8883288096161303,
      "eval_f1": 0.16034985422740525,
      "eval_loss": 0.3358159363269806,
      "eval_mcc": 0.17495541464580086,
      "eval_precision": 0.4700854700854701,
      "eval_recall": 0.09666080843585237,
      "eval_runtime": 9.9229,
      "eval_samples_per_second": 519.808,
      "eval_steps_per_second": 16.326,
      "step": 1290
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9260676503181458,
      "learning_rate": 0.00040477286728998583,
      "loss": 0.3287,
      "step": 1935
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8896859247770453,
      "eval_f1": 0.0,
      "eval_loss": 0.31568145751953125,
      "eval_mcc": 0.0,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 9.9841,
      "eval_samples_per_second": 516.623,
      "eval_steps_per_second": 16.226,
      "step": 1935
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.0631103515625,
      "learning_rate": 0.000346948171962845,
      "loss": 0.3215,
      "step": 2580
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8912369135323769,
      "eval_f1": 0.15639097744360905,
      "eval_loss": 0.3260652720928192,
      "eval_mcc": 0.18961519340791813,
      "eval_precision": 0.5416666666666666,
      "eval_recall": 0.0913884007029877,
      "eval_runtime": 9.9378,
      "eval_samples_per_second": 519.03,
      "eval_steps_per_second": 16.301,
      "step": 2580
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.190526008605957,
      "learning_rate": 0.0002891234766357042,
      "loss": 0.3187,
      "step": 3225
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8906552927491276,
      "eval_f1": 0.05369127516778523,
      "eval_loss": 0.3109687268733978,
      "eval_mcc": 0.11167217315148596,
      "eval_precision": 0.5925925925925926,
      "eval_recall": 0.028119507908611598,
      "eval_runtime": 9.9324,
      "eval_samples_per_second": 519.31,
      "eval_steps_per_second": 16.31,
      "step": 3225
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6695261597633362,
      "learning_rate": 0.00023129878130856335,
      "loss": 0.3131,
      "step": 3870
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.892594028693292,
      "eval_f1": 0.13975155279503107,
      "eval_loss": 0.30786898732185364,
      "eval_mcc": 0.18986913581302958,
      "eval_precision": 0.6,
      "eval_recall": 0.07908611599297012,
      "eval_runtime": 9.931,
      "eval_samples_per_second": 519.384,
      "eval_steps_per_second": 16.313,
      "step": 3870
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.1901545524597168,
      "learning_rate": 0.0001734740859814225,
      "loss": 0.3069,
      "step": 4515
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8924001550988755,
      "eval_f1": 0.2280945757997218,
      "eval_loss": 0.3023030459880829,
      "eval_mcc": 0.24105558868602414,
      "eval_precision": 0.5466666666666666,
      "eval_recall": 0.14411247803163443,
      "eval_runtime": 9.9209,
      "eval_samples_per_second": 519.913,
      "eval_steps_per_second": 16.329,
      "step": 4515
    },
    {
      "epoch": 8.0,
      "grad_norm": 7.957598686218262,
      "learning_rate": 0.00011564939065428167,
      "loss": 0.3002,
      "step": 5160
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8941450174486235,
      "eval_f1": 0.2132564841498559,
      "eval_loss": 0.3005247414112091,
      "eval_mcc": 0.24231018721764655,
      "eval_precision": 0.592,
      "eval_recall": 0.13005272407732865,
      "eval_runtime": 9.9245,
      "eval_samples_per_second": 519.726,
      "eval_steps_per_second": 16.323,
      "step": 5160
    }
  ],
  "logging_steps": 500,
  "max_steps": 6450,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 12695540260800.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.976928179899354,
    "learning_rate": 0.0005782469532714083,
    "num_train_epochs": 10,
    "temperature": 24
  }
}