metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: label
dtype: bool
splits:
- name: train
num_bytes: 13635269
num_examples: 4978
download_size: 6181968
dataset_size: 13635269
configs:
- config_name: default
data_files:
- split: train
path: data/train-*