metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: api
dtype: string
splits:
- name: train
num_bytes: 2002152
num_examples: 1357
download_size: 791426
dataset_size: 2002152
configs:
- config_name: default
data_files:
- split: train
path: data/train-*