---
dataset_info:
  features:
  - name: title
    dtype: string
  - name: abstract
    dtype: string
  - name: embeddings
    sequence: float32
  - name: id
    dtype: int64
  splits:
  - name: train
    num_bytes: 31471307429
    num_examples: 6758661
  download_size: 34423184917
  dataset_size: 31471307429
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---