metadata
dataset_info:
features:
- name: text
dtype: string
- name: doc_id
dtype: string
- name: meta
dtype: string
- name: quality_signals
dtype: string
splits:
- name: train
num_bytes: 25875844825.731113
num_examples: 2210971
download_size: 9343172262
dataset_size: 25875844825.731113
configs:
- config_name: default
data_files:
- split: train
path: data/train-*