metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
splits:
- name: train
num_bytes: 155519898
num_examples: 100000
- name: validation
num_bytes: 10037679
num_examples: 6447
- name: test
num_bytes: 10217787
num_examples: 6553
download_size: 106852761
dataset_size: 175775364
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*