metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 3220204.6999468934
num_examples: 1506
- name: test
num_bytes: 404129.2750929368
num_examples: 189
- name: valid
num_bytes: 401991.0249601699
num_examples: 188
download_size: 726354
dataset_size: 4026325.0000000005
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*