processing_summary.json

{
  "selected_repo_id": "agentica-org/DeepScaleR-Preview-Dataset",
  "source_file": "deepscaler_preview.json",
  "total_input_records": 40543,
  "kept_records": 39129,
  "dropped_records": 1414,
  "drop_breakdown": {
    "missing_problem": 8,
    "missing_answer": 46,
    "duplicate_problem": 1360
  },
  "output_rows": 39129
}