---
dataset_info:
  features:
  - name: label
    dtype:
      class_label:
        names:
          '0': negative
          '1': positive
  - name: input_mask
    sequence: int64
  - name: input_ids
    sequence: int32
  - name: actual_token_len
    sequence: int64
  splits:
  - name: train
    num_bytes: 11160000000
    num_examples: 3600000
  - name: test
    num_bytes: 1240000000
    num_examples: 400000
  download_size: 862017650
  dataset_size: 12400000000
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---