metadata
dataset_info:
features:
- name: type
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 109874
num_examples: 38
download_size: 58702
dataset_size: 109874
configs:
- config_name: default
data_files:
- split: train
path: data/train-*