{
  "teacher_model": "gpt2-large",
  "student_model": "distilgpt2",
  "tokenizer_model": "gpt2",
  "batch_size": 4,
  "max_length": 256,
  "num_epochs": 1,
  "learning_rate": 5e-05,
  "distillation_alpha": 0.7,
  "student_lm_beta": 0.3,
  "dataset_file": "/home/jayanthram.k/Distillation/clean_dataset_with_no_outliers.csv",
  "train_samples": 43572,
  "val_samples": 4842,
  "eval_samples": 100,
  "best_validation_loss": 2.3524647863515638,
  "evaluation_metrics": {
    "rouge1_teacher_student_before": 0.32189833105454313,
    "rouge2_teacher_student_before": 0.07253675599508494,
    "rougeL_teacher_student_before": 0.19562937213425652,
    "cosine_similarity_teacher_student_before": 0.44240760803222656,
    "rouge1_teacher_student_after": 0.3346180933711459,
    "rouge2_teacher_student_after": 0.08527329173516719,
    "rougeL_teacher_student_after": 0.20919089238473532,
    "cosine_similarity_teacher_student_after": 0.5256437659263611,
    "rouge1_student_before_student_after": 0.3137800181943293,
    "rouge2_student_before_student_after": 0.07712439299277635,
    "rougeL_student_before_student_after": 0.18683832802096778,
    "cosine_similarity_student_before_student_after": 0.4463849663734436,
    "avg_teacher_time": 1.2837520718574524,
    "avg_student_before_time": 0.4636122870445252,
    "avg_student_after_time": 0.46685707569122314
  }
}