metadata
dataset_info:
features:
- name: paragraphs
list:
- name: context
dtype: string
- name: qas
list:
- name: answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: id
dtype: string
- name: is_impossible
dtype: bool
- name: plausible_answers
list:
- name: answer_start
dtype: int64
- name: text
dtype: string
- name: question
dtype: string
- name: title
dtype: string
splits:
- name: train
num_bytes: 26130642.220963173
num_examples: 352
- name: test
num_bytes: 6658929.86199095
num_examples: 89
- name: validation
num_bytes: 3199100
num_examples: 35
download_size: 17910652
dataset_size: 35988672.08295412
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*