metadata
dataset_info:
features:
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 81800000
num_examples: 50000
download_size: 390890
dataset_size: 81800000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*