metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 1603533
num_examples: 4634
- name: valid
num_bytes: 534185
num_examples: 1545
- name: test
num_bytes: 535291
num_examples: 1545
download_size: 753817
dataset_size: 2673009
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: valid
path: data/valid-*
- split: test
path: data/test-*