metadata
dataset_info:
features:
- name: bytes
dtype: string
- name: path
dtype: string
- name: index
dtype: int64
- name: question
dtype: string
- name: multi-choice options
sequence: string
- name: answer
dtype: string
- name: category
dtype: string
- name: l2-category
dtype: string
splits:
- name: train
num_bytes: 34872184498
num_examples: 23609
download_size: 30887373363
dataset_size: 34872184498
configs:
- config_name: default
data_files:
- split: train
path: data/train-*