metadata
dataset_info:
features:
- name: text
dtype: string
- name: timestamp
dtype: string
- name: url
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 5257951783
num_examples: 1000000
download_size: 3117671463
dataset_size: 5257951783
configs:
- config_name: default
data_files:
- split: train
path: data/train-*