{
  "best_metric": 0.693069306930693,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-44/checkpoint-1188",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1485,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.536985993385315,
      "learning_rate": 0.0001924744581380813,
      "loss": 0.6378,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5326732673267327,
      "eval_f1": 0.39175257731958757,
      "eval_loss": 0.6261491775512695,
      "eval_mcc": 0.07262518750715272,
      "eval_precision": 0.5588235294117647,
      "eval_recall": 0.30158730158730157,
      "eval_runtime": 0.9367,
      "eval_samples_per_second": 539.146,
      "eval_steps_per_second": 17.082,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.5330690145492554,
      "learning_rate": 0.00015397956651046503,
      "loss": 0.6119,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6554455445544555,
      "eval_f1": 0.6824817518248175,
      "eval_loss": 0.5914366841316223,
      "eval_mcc": 0.31595686996005884,
      "eval_precision": 0.6317567567567568,
      "eval_recall": 0.7420634920634921,
      "eval_runtime": 0.9379,
      "eval_samples_per_second": 538.418,
      "eval_steps_per_second": 17.059,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.7286975383758545,
      "learning_rate": 0.00011548467488284878,
      "loss": 0.5882,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6772277227722773,
      "eval_f1": 0.7296849087893864,
      "eval_loss": 0.5896957516670227,
      "eval_mcc": 0.38579394462448086,
      "eval_precision": 0.6267806267806267,
      "eval_recall": 0.873015873015873,
      "eval_runtime": 0.9307,
      "eval_samples_per_second": 542.595,
      "eval_steps_per_second": 17.191,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.8389002084732056,
      "learning_rate": 7.698978325523251e-05,
      "loss": 0.5675,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.693069306930693,
      "eval_f1": 0.7207207207207208,
      "eval_loss": 0.5686942934989929,
      "eval_mcc": 0.3945060521718806,
      "eval_precision": 0.6600660066006601,
      "eval_recall": 0.7936507936507936,
      "eval_runtime": 0.9328,
      "eval_samples_per_second": 541.39,
      "eval_steps_per_second": 17.153,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.604902744293213,
      "learning_rate": 3.849489162761626e-05,
      "loss": 0.5527,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.691089108910891,
      "eval_f1": 0.7224199288256227,
      "eval_loss": 0.5650379657745361,
      "eval_mcc": 0.3929543844736987,
      "eval_precision": 0.6548387096774193,
      "eval_recall": 0.8055555555555556,
      "eval_runtime": 0.9326,
      "eval_samples_per_second": 541.477,
      "eval_steps_per_second": 17.156,
      "step": 1485
    }
  ],
  "logging_steps": 500,
  "max_steps": 1782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 3653506329900.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8986952116548445,
    "learning_rate": 0.00023096934976569756,
    "num_train_epochs": 6,
    "temperature": 24
  }
}