metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 7084750.389650392
num_examples: 5194
- name: test
num_bytes: 886616.8181118127
num_examples: 650
- name: valid
num_bytes: 885252.7922377945
num_examples: 649
download_size: 1025467
dataset_size: 8856620
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*