---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: topic
    dtype: string
  - name: title
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: special_tokens_mask
    sequence: int8
  - name: chunk
    dtype: int64
  splits:
  - name: train
    num_bytes: 1257397482
    num_examples: 17237
  download_size: 419630414
  dataset_size: 1257397482
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---