metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: api
dtype: string
splits:
- name: train
num_bytes: 13058410
num_examples: 4724
- name: test
num_bytes: 1817463
num_examples: 410
download_size: 1575537
dataset_size: 14875873
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*