---
dataset_info:
  features:
  - name: tokens
    sequence: int64
  - name: labels
    sequence: int64
  - name: loss_mask
    sequence: int64
  - name: position_ids
    sequence: int64
  splits:
  - name: train
    num_bytes: 381704960
    num_examples: 1456
  - name: validation
    num_bytes: 2097280
    num_examples: 8
  - name: test
    num_bytes: 1048640
    num_examples: 4
  download_size: 12661098
  dataset_size: 384850880
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---