metadata
dataset_info:
features:
- name: text
dtype: string
- name: metadata
dtype: string
splits:
- name: train
num_bytes: 44266046418
num_examples: 2875658
download_size: 13512248530
dataset_size: 44266046418
configs:
- config_name: default
data_files:
- split: train
path: data/train-*