metadata
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: evaluation
path: data/evaluation-*
- split: test
path: data/test-*
- split: train_full
path: data/train_full-*
dataset_info:
features:
- name: text1
dtype: string
- name: text2
dtype: string
splits:
- name: train
num_bytes: 246977207
num_examples: 900000
- name: evaluation
num_bytes: 27414347
num_examples: 100000
- name: test
num_bytes: 27471369
num_examples: 100000
- name: train_full
num_bytes: 274391554
num_examples: 1000000
download_size: 189206059
dataset_size: 576254477