---
dataset_info:
  features:
  - name: label
    dtype: int64
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  splits:
  - name: train
    num_bytes: 26306985
    num_examples: 20000
  - name: test
    num_bytes: 32420025
    num_examples: 25000
  - name: dev
    num_bytes: 6636740
    num_examples: 5000
  download_size: 28273093
  dataset_size: 65363750
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
  - split: dev
    path: data/dev-*
---