metadata
dataset_info:
features:
- name: raw_content
dtype: string
- name: doc_id
dtype: string
- name: meta
dtype: string
- name: quality_signals
dtype: string
splits:
- name: train
num_bytes: 104068894016
num_examples: 10000000
download_size: 0
dataset_size: 104068894016