---
dataset_info:
  features:
  - name: tokens
    sequence: int64
  - name: targets
    sequence: int64
  splits:
  - name: train
    num_bytes: 49288800
    num_examples: 3300
  download_size: 3044454
  dataset_size: 49288800
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---