metadata
dataset_info:
features:
- name: category_original
dtype: string
- name: prompt_original
dtype: string
- name: prompt_original_copy
dtype: string
splits:
- name: train
num_bytes: 76001
num_examples: 129
download_size: 46574
dataset_size: 76001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*