metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 1415916
num_examples: 1859
- name: valid
num_bytes: 1440885
num_examples: 1860
- name: test
num_bytes: 1421070
num_examples: 1849
download_size: 1533748
dataset_size: 4277871
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*