metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: chosen
dtype: string
- name: hash
dtype: string
splits:
- name: train
num_bytes: 26429481
num_examples: 23332
download_size: 14742258
dataset_size: 26429481
configs:
- config_name: default
data_files:
- split: train
path: data/train-*