metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 253172
num_examples: 2379
download_size: 150776
dataset_size: 253172
configs:
- config_name: default
data_files:
- split: train
path: data/train-*