metadata
dataset_info:
features:
- name: instruction
dtype: string
- name: context
dtype: string
- name: response
dtype: string
splits:
- name: train
num_bytes: 5091
num_examples: 12
- name: validation
num_bytes: 1697
num_examples: 4
download_size: 11272
dataset_size: 6788
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*