metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: label
dtype: bool
- name: source
dtype: string
- name: base_prompt
dtype: string
- name: index
dtype: int64
splits:
- name: train
num_bytes: 964979156
num_examples: 1147951
download_size: 252336309
dataset_size: 964979156
configs:
- config_name: default
data_files:
- split: train
path: data/train-*