---
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: timestamp[us]
  - name: url
    dtype: string
  - name: doc_hash
    dtype: string
  splits:
  - name: train
    num_bytes: 23070
    num_examples: 21
  download_size: 25466
  dataset_size: 23070
configs:
- config_name: default
  data_files:
  - split: train
    path: 1f9bd48d4907bb9d/train-*
---