metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 4120244
num_examples: 3597
download_size: 2491202
dataset_size: 4120244
configs:
- config_name: default
data_files:
- split: train
path: data/train-*