{
  "best_metric": 0.763921568627451,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-16/checkpoint-800",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.7216335535049438,
      "learning_rate": 0.00033371467549721734,
      "loss": 0.4146,
      "step": 160
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6870588235294117,
      "eval_f1": 0.1971830985915493,
      "eval_loss": 0.37955978512763977,
      "eval_mcc": 0.18019685365777874,
      "eval_precision": 0.6805555555555556,
      "eval_recall": 0.11529411764705882,
      "eval_runtime": 1.871,
      "eval_samples_per_second": 681.446,
      "eval_steps_per_second": 21.379,
      "step": 160
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9292495250701904,
      "learning_rate": 0.00028604115042618627,
      "loss": 0.3901,
      "step": 320
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7137254901960784,
      "eval_f1": 0.33515482695810567,
      "eval_loss": 0.3745056092739105,
      "eval_mcc": 0.2844985686324911,
      "eval_precision": 0.7419354838709677,
      "eval_recall": 0.2164705882352941,
      "eval_runtime": 1.8722,
      "eval_samples_per_second": 681.009,
      "eval_steps_per_second": 21.365,
      "step": 320
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.6019469499588013,
      "learning_rate": 0.00023836762535515526,
      "loss": 0.3811,
      "step": 480
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7372549019607844,
      "eval_f1": 0.5007451564828614,
      "eval_loss": 0.37347400188446045,
      "eval_mcc": 0.3626013921763011,
      "eval_precision": 0.6829268292682927,
      "eval_recall": 0.3952941176470588,
      "eval_runtime": 1.8665,
      "eval_samples_per_second": 683.081,
      "eval_steps_per_second": 21.43,
      "step": 480
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7646651268005371,
      "learning_rate": 0.0001906941002841242,
      "loss": 0.3758,
      "step": 640
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7309803921568627,
      "eval_f1": 0.40963855421686746,
      "eval_loss": 0.3691938519477844,
      "eval_mcc": 0.3401760368221827,
      "eval_precision": 0.7628205128205128,
      "eval_recall": 0.28,
      "eval_runtime": 1.8659,
      "eval_samples_per_second": 683.299,
      "eval_steps_per_second": 21.437,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9724321365356445,
      "learning_rate": 0.00014302057521309314,
      "loss": 0.3704,
      "step": 800
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.763921568627451,
      "eval_f1": 0.579020979020979,
      "eval_loss": 0.3683825731277466,
      "eval_mcc": 0.4379211108776948,
      "eval_precision": 0.7137931034482758,
      "eval_recall": 0.48705882352941177,
      "eval_runtime": 1.8659,
      "eval_samples_per_second": 683.325,
      "eval_steps_per_second": 21.438,
      "step": 800
    }
  ],
  "logging_steps": 500,
  "max_steps": 1280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1458777189600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.640212771663602,
    "learning_rate": 0.0003813882005682484,
    "num_train_epochs": 8,
    "temperature": 30
  }
}