metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 5112013
num_examples: 3674
download_size: 486210
dataset_size: 5112013
configs:
- config_name: default
data_files:
- split: train
path: data/train-*