metadata
dataset_info:
features:
- name: system
dtype: string
- name: prompt
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 7058529
num_examples: 870
download_size: 733943
dataset_size: 7058529
configs:
- config_name: default
data_files:
- split: train
path: data/train-*