---
dataset_info:
  features:
  - name: example_idx
    dtype: int32
  - name: text
    dtype: string
  - name: normalized_text
    dtype: string
  - name: token_ids_utf8_bos
    list: int32
  - name: text_encoder_hidden_states
    list:
      list: float32
  splits:
  - name: train
    num_bytes: 117702947
    num_examples: 5
  download_size: 44205672
  dataset_size: 117702947
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---