metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 2548717
num_examples: 4496
download_size: 390773
dataset_size: 2548717
configs:
- config_name: default
data_files:
- split: train
path: data/train-*