metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: text
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 46660
num_examples: 80
- name: test
num_bytes: 11665
num_examples: 20
download_size: 31306
dataset_size: 58325
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*