---
dataset_info:
  features:
  - name: quaternion
    list: string
  - name: best
    dtype: string
  - name: worst
    dtype: string
  - name: annotator_llm
    dtype: string
  splits:
  - name: train
    num_bytes: 9513744
    num_examples: 9799
  download_size: 3677784
  dataset_size: 9513744
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---