metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 8054650
num_examples: 4670
download_size: 1077634
dataset_size: 8054650
configs:
- config_name: default
data_files:
- split: train
path: data/train-*