metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: api
dtype: string
splits:
- name: train
num_bytes: 19992016
num_examples: 3439
- name: test
num_bytes: 779714
num_examples: 248
download_size: 2771494
dataset_size: 20771730
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*