metadata
dataset_info:
features:
- name: id
dtype: int64
- name: instruction
dtype: string
- name: options
struct:
- name: label
sequence: string
- name: text
sequence: string
- name: answerKey
dtype: string
splits:
- name: train
num_bytes: 27298834
num_examples: 37116
download_size: 3568963
dataset_size: 27298834
configs:
- config_name: default
data_files:
- split: train
path: data/train-*