dataset_stats.json

{
  "total_examples": 100000,
  "splits": {
    "train": 89916,
    "validation": 5104,
    "test": 4980
  },
  "question_type_distribution": {
    "general_natal": 25000,
    "career": 25000,
    "marriage_timing": 25000,
    "sub_lord_reading": 25000
  },
  "avg_citations_per_example": 4.097,
  "examples_with_zero_citations": 0
}