metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: choices
dtype: string
- name: answerKey
dtype: string
- name: right_context
dtype: string
- name: eval
dtype: bool
- name: substituted
dtype: bool
- name: correct_answer
dtype: string
splits:
- name: train
num_bytes: 4315810
num_examples: 797
download_size: 1047260
dataset_size: 4315810
configs:
- config_name: default
data_files:
- split: train
path: data/train-*