---
dataset_info:
  features:
  - name: source
    dtype: string
  - name: target
    dtype: string
  - name: target_backto_source
    dtype: string
  - name: raw_target
    list:
    - name: generated_text
      dtype: string
  - name: raw_target_backto_source
    list:
    - name: generated_text
      dtype: string
  - name: prompt
    dtype: string
  - name: reverse_prompt
    dtype: string
  - name: source_langid
    dtype: string
  - name: target_langid
    dtype: string
  - name: target_backto_source_langid
    dtype: string
  - name: doc_id
    dtype: int64
  - name: sent_id
    dtype: int64
  - name: timestamp
    dtype: string
  - name: url
    dtype: string
  - name: doc_hash
    dtype: string
  splits:
  - name: train
    num_bytes: 4420
    num_examples: 5
  download_size: 12105
  dataset_size: 4420
configs:
- config_name: default
  data_files:
  - split: train
    path: a46dd2d9665b7264/train-*
---