metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: api
dtype: string
splits:
- name: train
num_bytes: 40858259
num_examples: 3913
- name: test
num_bytes: 3647777
num_examples: 728
download_size: 8068002
dataset_size: 44506036
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*