metadata
dataset_info:
features:
- name: id
dtype: string
- name: question
dtype: string
- name: choices
struct:
- name: label
list: string
- name: text
list: string
- name: answerKey
dtype: string
- name: instruction_text
dtype: string
- name: dataset_name
dtype: string
- name: id_num
dtype: int64
- name: seen
dtype: bool
splits:
- name: train
num_bytes: 803824
num_examples: 1000
- name: test
num_bytes: 84325
num_examples: 100
download_size: 391357
dataset_size: 888149
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*