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: 14980980.653468275
num_examples: 26019
- name: validation
num_bytes: 1665129.176748924
num_examples: 2892
- name: test
num_bytes: 876323.1697828015
num_examples: 1522
download_size: 1948135
dataset_size: 17522433
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*