metadata
dataset_info:
features:
- name: id
dtype: string
- name: prompt
dtype: string
- name: response
dtype: string
- name: categories
list: string
- name: category_count
dtype: int64
- name: category_string
dtype: string
- name: prompt_label
dtype: string
- name: response_label
dtype: string
- name: prompt_label_source
dtype: string
- name: response_label_source
dtype: string
- name: has_needs_caution
dtype: bool
splits:
- name: train
num_bytes: 1580319
num_examples: 2292
download_size: 794750
dataset_size: 1580319
configs:
- config_name: default
data_files:
- split: train
path: data/train-*