metadata
dataset_info:
features:
- name: source
dtype: string
- name: text
dtype: string
- name: prompt
dtype: string
splits:
- name: train
num_bytes: 290333249
num_examples: 281057
download_size: 153072934
dataset_size: 290333249
configs:
- config_name: default
data_files:
- split: train
path: data/train-*