metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 1899249
num_examples: 1869
- name: valid
num_bytes: 1859682
num_examples: 1873
- name: test
num_bytes: 1918377
num_examples: 1860
download_size: 1550655
dataset_size: 5677308
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*