{
  "best_metric": 0.6237623762376238,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/checkpoint-1782",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1782,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.202705979347229,
      "learning_rate": 6.504485936113838e-05,
      "loss": 0.5495,
      "step": 297
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5069306930693069,
      "eval_f1": 0.10108303249097472,
      "eval_loss": 0.5468167662620544,
      "eval_mcc": 0.027838098756040194,
      "eval_precision": 0.56,
      "eval_recall": 0.05555555555555555,
      "eval_runtime": 0.9316,
      "eval_samples_per_second": 542.063,
      "eval_steps_per_second": 17.174,
      "step": 297
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8953370451927185,
      "learning_rate": 5.4204049467615325e-05,
      "loss": 0.54,
      "step": 594
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.594059405940594,
      "eval_f1": 0.5858585858585857,
      "eval_loss": 0.5394836068153381,
      "eval_mcc": 0.18817791261380143,
      "eval_precision": 0.5967078189300411,
      "eval_recall": 0.5753968253968254,
      "eval_runtime": 0.9335,
      "eval_samples_per_second": 540.967,
      "eval_steps_per_second": 17.14,
      "step": 594
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.554856777191162,
      "learning_rate": 4.336323957409226e-05,
      "loss": 0.5316,
      "step": 891
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5861386138613861,
      "eval_f1": 0.6365217391304347,
      "eval_loss": 0.5389003157615662,
      "eval_mcc": 0.17998847484487682,
      "eval_precision": 0.56656346749226,
      "eval_recall": 0.7261904761904762,
      "eval_runtime": 0.9343,
      "eval_samples_per_second": 540.483,
      "eval_steps_per_second": 17.124,
      "step": 891
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0591225624084473,
      "learning_rate": 3.252242968056919e-05,
      "loss": 0.527,
      "step": 1188
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6118811881188119,
      "eval_f1": 0.6287878787878789,
      "eval_loss": 0.5312101244926453,
      "eval_mcc": 0.22492336097439145,
      "eval_precision": 0.6014492753623188,
      "eval_recall": 0.6587301587301587,
      "eval_runtime": 0.9347,
      "eval_samples_per_second": 540.304,
      "eval_steps_per_second": 17.119,
      "step": 1188
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.7271960973739624,
      "learning_rate": 2.168161978704613e-05,
      "loss": 0.5226,
      "step": 1485
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6138613861386139,
      "eval_f1": 0.6183953033268101,
      "eval_loss": 0.5254874229431152,
      "eval_mcc": 0.22784970249121222,
      "eval_precision": 0.61003861003861,
      "eval_recall": 0.626984126984127,
      "eval_runtime": 0.9248,
      "eval_samples_per_second": 546.078,
      "eval_steps_per_second": 17.301,
      "step": 1485
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.6252127885818481,
      "learning_rate": 1.0840809893523065e-05,
      "loss": 0.5188,
      "step": 1782
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6237623762376238,
      "eval_f1": 0.6468401486988847,
      "eval_loss": 0.5257902145385742,
      "eval_mcc": 0.24999798285978939,
      "eval_precision": 0.6083916083916084,
      "eval_recall": 0.6904761904761905,
      "eval_runtime": 0.9286,
      "eval_samples_per_second": 543.832,
      "eval_steps_per_second": 17.23,
      "step": 1782
    }
  ],
  "logging_steps": 500,
  "max_steps": 2079,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 4384207595880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7404813991868276,
    "learning_rate": 7.588566925466145e-05,
    "num_train_epochs": 7,
    "temperature": 47
  }
}