metadata
dataset_info:
features:
- name: example_id
dtype: int64
- name: category
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: ans0
dtype: string
- name: ans1
dtype: string
- name: ans2
dtype: string
- name: label
dtype: int64
- name: unknown_num
dtype: int64
splits:
- name: train
num_bytes: 6692254
num_examples: 40000
download_size: 479719
dataset_size: 6692254
configs:
- config_name: default
data_files:
- split: train
path: data/train-*