Datasets:
metadata
dataset_info:
features:
- name: response
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 17694132
num_examples: 46283
download_size: 11503545
dataset_size: 17694132
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- nb