---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: token_length
    dtype: int64
  - name: system_prompt
    dtype: string
  - name: generations
    sequence: string
  - name: distilabel_metadata
    struct:
    - name: raw_output_ultra_feedback_0
      dtype: string
  - name: model_names
    sequence: string
  - name: ratings
    sequence: int64
  - name: rationales
    sequence: string
  - name: model_name
    dtype: string
  splits:
  - name: train
    num_bytes: 1782844
    num_examples: 200
  download_size: 937298
  dataset_size: 1782844
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---