metadata
dataset_info:
features:
- name: knowledge
dtype: string
- name: query
dtype: string
- name: response
dtype: string
- name: label
dtype: string
- name: reason
dtype: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 92743601
num_examples: 78890
- name: test
num_bytes: 13225722
num_examples: 13960
download_size: 56496213
dataset_size: 105969323
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*