metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
dataset_info:
features:
- name: id
dtype: string
- name: language
dtype: string
- name: document_title
dtype: string
- name: passage_text
dtype: string
- name: question_text
dtype: string
- name: answers
sequence:
- name: text
dtype: string
- name: start_byte
dtype: int32
- name: limit_byte
dtype: int32
splits:
- name: train
num_bytes: 16692327
num_examples: 14805
- name: validation
num_bytes: 1063786
num_examples: 921
download_size: 9214756
dataset_size: 17756113