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: 25604886.41862452
num_examples: 24433
- name: test
num_bytes: 3143889.7906877403
num_examples: 3000
- name: validation
num_bytes: 3143889.7906877403
num_examples: 3000
download_size: 2927041
dataset_size: 31892666
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*