metadata
dataset_info:
features:
- name: bad_answer
dtype: string
- name: percent
dtype: int64
- name: question
dtype: string
splits:
- name: train
num_bytes: 13321627
num_examples: 6000
download_size: 3940300
dataset_size: 13321627
configs:
- config_name: default
data_files:
- split: train
path: data/train-*