metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 2778281
num_examples: 539
download_size: 575887
dataset_size: 2778281
configs:
- config_name: default
data_files:
- split: train
path: data/train-*