metadata
language:
- en
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: 23339845.7129001
num_examples: 31734
- name: validation
num_bytes: 2593316.190322233
num_examples: 3526
- name: test
num_bytes: 1365058.09677767
num_examples: 1856
download_size: 3585201
dataset_size: 27298220
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*