metadata
dataset_info:
features:
- name: sha
dtype: string
- name: text
dtype: string
- name: id
dtype: string
- name: tags
sequence: string
- name: created_at
dtype: string
- name: metadata
dtype: string
- name: last_modified
dtype: string
- name: arxiv
sequence: string
- name: languages
sequence: string
- name: tags_str
dtype: string
- name: text_str
dtype: string
- name: text_lists
sequence: string
- name: processed_texts
sequence: string
- name: tokens_length
sequence: int64
- name: input_texts
sequence: string
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 750959701
num_examples: 52904
download_size: 331683974
dataset_size: 750959701
configs:
- config_name: default
data_files:
- split: train
path: data/train-*