metadata
dataset_info:
features:
- name: context
dtype: string
- name: question
dtype: string
- name: answers
sequence:
- name: answer_start
dtype: int32
- name: text
dtype: string
- name: id
dtype: string
- name: prompt
dtype: string
splits:
- name: test
num_bytes: 335265
num_examples: 200
download_size: 186354
dataset_size: 335265
configs:
- config_name: default
data_files:
- split: test
path: data/test-*