metadata
dataset_info:
features:
- name: knowledge
dtype: string
- name: question
dtype: string
- name: right_answer
dtype: string
- name: hallucinated_answer
dtype: string
splits:
- name: train
num_bytes: 5453425
num_examples: 10000
download_size: 3730403
dataset_size: 5453425
configs:
- config_name: default
data_files:
- split: train
path: data/train-*