metadata
dataset_info:
features:
- name: passage_content
dtype: string
- name: passage_id
dtype: string
splits:
- name: train_passages
num_bytes: 42054675
num_examples: 112724
- name: valid_passages
num_bytes: 61645843
num_examples: 168306
- name: test_passages
num_bytes: 61645843
num_examples: 168306
download_size: 98987092
dataset_size: 165346361
configs:
- config_name: default
data_files:
- split: train_passages
path: data/train_passages-*
- split: valid_passages
path: data/valid_passages-*
- split: test_passages
path: data/test_passages-*