metadata
dataset_info:
features:
- name: id
dtype: string
- name: prompt
list:
- name: content
dtype: string
- name: role
dtype: string
- name: constraint
list:
list:
- name: id
dtype: string
- name: placeholder
dtype: string
- name: persona
dtype: string
splits:
- name: train
num_bytes: 6676169
num_examples: 12389
download_size: 3539832
dataset_size: 6676169
configs:
- config_name: default
data_files:
- split: train
path: data/train-*