metadata
dataset_info:
features:
- name: text
dtype: string
- name: writing_style_average
dtype: float64
- name: facts_and_trivia_average
dtype: float64
- name: educational_value_average
dtype: float64
- name: required_expertise_average
dtype: float64
- name: writing_style_chunks
sequence: float64
- name: facts_and_trivia_chunks
sequence: float64
- name: educational_value_chunks
sequence: float64
- name: required_expertise_chunks
sequence: float64
- name: length
dtype: int64
- name: chunk_lengths
sequence: int64
- name: input_ids
sequence: int32
- name: document_index
dtype: int64
- name: document_position
dtype: int64
- name: source_domain
dtype: string
- name: cluster_id
dtype: int64
- name: cluster_no
dtype: int64
splits:
- name: train
num_bytes: 8042646032
num_examples: 1000000
download_size: 4212824141
dataset_size: 8042646032
configs:
- config_name: default
data_files:
- split: train
path: data/train-*