---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: pdf_path
    dtype: string
  - name: pdf_name
    dtype: string
  - name: year
    dtype: string
  - name: start_page
    dtype: int64
  - name: end_page
    dtype: int64
  - name: chunk_idx
    dtype: int64
  - name: start_tok
    dtype: int64
  - name: end_tok
    dtype: int64
  - name: text
    dtype: string
  - name: generated_text
    dtype: string
  - name: parsed_question_list
    list:
    - name: answer
      dtype: string
    - name: question
      dtype: string
  splits:
  - name: train
    num_bytes: 72873825
    num_examples: 11665
  download_size: 27256868
  dataset_size: 72873825
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---