metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: completion
dtype: string
- name: dataset
dtype: string
splits:
- name: train
num_bytes: 10098690
num_examples: 28069
- name: validation
num_bytes: 1895607
num_examples: 5263
- name: test
num_bytes: 627329
num_examples: 1755
download_size: 4772208
dataset_size: 12621626
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*