metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 3909355
num_examples: 3906
download_size: 1410252
dataset_size: 3909355
configs:
- config_name: default
data_files:
- split: train
path: data/train-*