metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 20407546.384733763
num_examples: 5198
- name: test
num_bytes: 2551924.807633118
num_examples: 650
- name: valid
num_bytes: 2551924.807633118
num_examples: 650
download_size: 2859145
dataset_size: 25511395.999999996
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*