metadata
dataset_info:
features:
- name: human
dtype: string
- name: gpt
dtype: string
- name: source
dtype: string
- name: score
dtype: float64
splits:
- name: train
num_bytes: 2934032
num_examples: 8036
download_size: 292187
dataset_size: 2934032
configs:
- config_name: default
data_files:
- split: train
path: data/train-*