The script that create datasaet see Kaggle notebook [Pre-process the dataset](https://www.kaggle.com/code/aisuko/pre-process-the-datasets)

---
dataset_info:
  features:
  - name: title
    dtype: string
  - name: question
    dtype: string
  - name: answer
    dtype: string
  - name: question_embedding
    sequence: float32
  splits:
  - name: validation
    num_bytes: 18179
    num_examples: 11
  download_size: 27832
  dataset_size: 18179
configs:
- config_name: default
  data_files:
  - split: validation
    path: data/validation-*
---