metadata
dataset_info:
features:
- name: content
dtype: string
- name: url
dtype: string
- name: timestamp
dtype: timestamp[s]
- name: dump
dtype: string
- name: segment
dtype: string
- name: image_urls
sequence:
sequence: string
splits:
- name: train
num_bytes: 2766953721769
num_examples: 968000015
download_size: 466888198663
dataset_size: 2766953721769