metadata
dataset_info:
features:
- name: weird_prompts
dtype: string
splits:
- name: train
num_bytes: 7102831
num_examples: 60481
download_size: 1800771
dataset_size: 7102831
configs:
- config_name: default
data_files:
- split: train
path: data/train-*