metadata
language:
- en
dataset_info:
features:
- name: Input
dtype: string
- name: Instruction
dtype: string
- name: Response
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 13145938
num_examples: 5000
download_size: 4962049
dataset_size: 13145938
configs:
- config_name: default
data_files:
- split: train
path: data/train-*