metadata
dataset_info:
features:
- name: text
dtype: string
- name: prompt
dtype: string
- name: text_token_length
dtype: int64
- name: seed_data
dtype: string
- name: format
dtype: string
- name: audience
dtype: string
splits:
- name: validation
num_bytes: 17024678
num_examples: 4000
download_size: 9263303
dataset_size: 17024678
configs:
- config_name: default
data_files:
- split: validation
path: data/validation-*