---
dataset_info:
  features:
  - name: original_data
    struct:
    - name: date
      dtype: string
    - name: object
      dtype: string
    - name: relation
      dtype: string
    - name: subject
      dtype: string
  - name: original_question
    dtype: string
  - name: original_answer
    dtype: string
  - name: question
    dtype: string
  - name: solution
    dtype: string
  - name: model_answer
    dtype: string
  - name: candidates
    sequence: string
  - name: reasoning_content
    dtype: string
  splits:
  - name: train
    num_bytes: 104478
    num_examples: 19
  - name: valid
    num_bytes: 23426
    num_examples: 4
  download_size: 112890
  dataset_size: 127904
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
---