metadata
dataset_info:
features:
- name: input
dtype: string
- name: output
dtype: int64
splits:
- name: train
num_bytes: 34303623.63657499
num_examples: 34410
- name: test
num_bytes: 4288700.635993769
num_examples: 4302
- name: validation
num_bytes: 4287703.727431241
num_examples: 4301
- name: discarded
num_bytes: 23203480
num_examples: 6987
download_size: 43060425
dataset_size: 66083508
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: validation
path: data/validation-*
- split: discarded
path: data/discarded-*