README.md

---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: question
    dtype: string
  - name: answers
    sequence: string
  - name: top1_answerable
    dtype: bool
  - name: top5_answerable
    dtype: bool
  - name: top5
    list:
    - name: atom_student_10
      dtype: string
    - name: atom_student_20
      dtype: string
    - name: atom_student_t5_proposition_v1
      dtype: string
    - name: atom_student_t5_proposition_v2
      dtype: string
    - name: atom_teacher
      dtype: string
    - name: bm25_score
      dtype: string
    - name: contents
      dtype: string
    - name: docID
      dtype: int64
    - name: has_answer
      dtype: bool
    - name: rank
      dtype: string
  splits:
  - name: nq
    num_bytes: 312433473
    num_examples: 3610
  download_size: 171117839
  dataset_size: 312433473
configs:
- config_name: default
  data_files:
  - split: nq
    path: data/nq-*
---