---
dataset_info:
  features:
  - name: conversation
    list:
    - name: content
      dtype: string
    - name: role
      dtype: string
  - name: chosen
    struct:
    - name: content
      struct:
      - name: cleaned_description
        dtype: string
      - name: cleaned_description_confidence_score
        dtype: int64
      - name: merchant_name
        dtype: string
      - name: merchant_name_confidence_score
        dtype: int64
      - name: transaction_description
        dtype: string
    - name: role
      dtype: string
  - name: rejected
    struct:
    - name: content
      struct:
      - name: cleaned_description
        dtype: string
      - name: cleaned_description_confidence_score
        dtype: int64
      - name: merchant_name
        dtype: string
      - name: merchant_name_confidence_score
        dtype: int64
      - name: transaction_description
        dtype: string
    - name: role
      dtype: string
  splits:
  - name: train
    num_bytes: 848655.7464454976
    num_examples: 675
  - name: test
    num_bytes: 212478.25355450236
    num_examples: 169
  download_size: 115571
  dataset_size: 1061134.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---