metadata
dataset_info:
features:
- name: text
dtype: string
- name: page_number
dtype: int64
- name: chunk_id
dtype: int64
- name: page_chunk
dtype: int64
- name: doc_id
dtype: string
- name: doc_num
dtype: int64
- name: doc_name
dtype: string
- name: n_tokens
dtype: int64
splits:
- name: train
num_bytes: 901176
num_examples: 925
download_size: 407846
dataset_size: 901176
configs:
- config_name: default
data_files:
- split: train
path: data/train-*