metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: api
dtype: string
splits:
- name: train
num_bytes: 4224309
num_examples: 2499
download_size: 726832
dataset_size: 4224309
configs:
- config_name: default
data_files:
- split: train
path: data/train-*