metadata
dataset_info:
features:
- name: template
dtype: string
- name: variable_text
dtype: string
- name: desired_text
dtype: string
- name: undesired_text
dtype: string
- name: story_type
dtype: string
- name: human_answer
dtype: string
splits:
- name: train
num_bytes: 32172
num_examples: 100
download_size: 24368
dataset_size: 32172
configs:
- config_name: default
data_files:
- split: train
path: data/train-*