metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
sequence: string
- name: id
dtype: string
splits:
- name: train
num_bytes: 7664553.422861277
num_examples: 4423
- name: test
num_bytes: 958285.7885693616
num_examples: 553
- name: valid
num_bytes: 958285.7885693616
num_examples: 553
download_size: 1690540
dataset_size: 9581125
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: valid
path: data/valid-*