---
dataset_info:
  features:
  - name: title
    dtype: string
  - name: topic
    dtype: string
  - name: text
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  - name: packed_length
    dtype: int64
  splits:
  - name: train
    num_bytes: 69601717.00154574
    num_examples: 78052
  - name: validation
    num_bytes: 3664139.9984542546
    num_examples: 4109
  download_size: 8705351
  dataset_size: 73265857.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
---