metadata
dataset_info:
features:
- name: text
dtype: string
- name: timestamp
dtype: string
- name: url
dtype: string
- name: perplexity_score
dtype: float64
- name: text_length
dtype: int64
- name: domain
dtype: 'null'
- name: dup_ratio
dtype: float64
- name: pairs
sequence:
sequence: int64
- name: repetitions
sequence: binary
- name: included_in_dedup
dtype: bool
- name: cluster
sequence: int64
splits:
- name: train
num_bytes: 959334093604
num_examples: 364868892
download_size: 586254318285
dataset_size: 959334093604