---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: input_ids
    sequence: int32
  - name: attention_mask
    sequence: int8
  - name: labels
    sequence: int64
  splits:
  - name: train
    num_bytes: 167752286.4
    num_examples: 800000
  - name: validation
    num_bytes: 20969035.8
    num_examples: 100000
  - name: test
    num_bytes: 20969035.8
    num_examples: 100000
  download_size: 106899710
  dataset_size: 209690358.00000003
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---