metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 1507005
num_examples: 1923
download_size: 941833
dataset_size: 1507005