---
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: 8171236
    num_examples: 808
  download_size: 1964368
  dataset_size: 8171236
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---