metadata
dataset_info:
features:
- name: name
dtype: string
- name: prompt
dtype: string
- name: cls_prompt
dtype: string
splits:
- name: train
num_bytes: 13704527
num_examples: 5000
download_size: 3414974
dataset_size: 13704527
configs:
- config_name: default
data_files:
- split: train
path: data/train-*