metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 501126
num_examples: 1625
- name: validation
num_bytes: 26013
num_examples: 86
download_size: 59970
dataset_size: 527139
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*