metadata
dataset_info:
features:
- name: question
dtype: string
- name: relevant_articles
list:
- name: article_content
dtype: string
- name: article_id
dtype: string
- name: docid
dtype: string
- name: document_id
dtype: string
- name: text
dtype: string
- name: title
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 70460543
num_examples: 12529
download_size: 10130728
dataset_size: 70460543
configs:
- config_name: default
data_files:
- split: train
path: data/train-*