metadata
dataset_info:
features:
- name: text
dtype: string
- name: instruction_response
list:
- name: A
dtype: string
- name: Q
dtype: string
splits:
- name: train
num_bytes: 12015
num_examples: 20
download_size: 12640
dataset_size: 12015
configs:
- config_name: default
data_files:
- split: train
path: data/train-*