metadata
dataset_info:
features:
- name: category
dtype: string
- name: prompt
list: string
- name: reference
list: string
- name: prompt_id
dtype: int64
- name: interactions
list:
- name: query
dtype: string
splits:
- name: train
num_bytes: 78933
num_examples: 80
download_size: 52977
dataset_size: 78933
configs:
- config_name: default
data_files:
- split: train
path: data/train-*