metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: history
dtype: string
- name: human_ref_A
dtype: string
splits:
- name: train
num_bytes: 27510
num_examples: 13
download_size: 30026
dataset_size: 27510
configs:
- config_name: default
data_files:
- split: train
path: data/train-*