---
dataset_info:
  features:
  - name: filename
    dtype: string
  - name: title
    dtype: string
  - name: abstract
    dtype: string
  - name: pred_lin
    sequence: string
  - name: prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 133287821
    num_examples: 13559
  download_size: 30776345
  dataset_size: 133287821
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---