---
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: 30928239
    num_examples: 3006
  download_size: 7559215
  dataset_size: 30928239
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---