metadata
dataset_info:
features:
- name: sentence1
sequence: string
- name: sentence2
sequence: string
- name: label
sequence: int64
splits:
- name: train
num_bytes: 2805906476
num_examples: 381015
download_size: 1419269500
dataset_size: 2805906476
configs:
- config_name: default
data_files:
- split: train
path: data/train-*