---
dataset_info:
  features:
  - name: speaker_id
    dtype: string
  - name: reference_text
    dtype: string
  - name: target_text
    dtype: string
  - name: target_tokens_1B
    sequence: int64
  - name: target_tokens_3B
    sequence: int64
  - name: target_tokens_8B
    sequence: int64
  splits:
  - name: train
    num_bytes: 593719
    num_examples: 37
  download_size: 255645
  dataset_size: 593719
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---