---
dataset_info:
  features:
  - name: tokens
    sequence: int64
  - name: targets
    sequence: int64
  splits:
  - name: train
    num_bytes: 49288800
    num_examples: 3300
  - name: eval
    num_bytes: 2748224
    num_examples: 184
  - name: test
    num_bytes: 1493600
    num_examples: 100
  download_size: 3341205
  dataset_size: 53530624
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: eval
    path: data/eval-*
  - split: test
    path: data/test-*
---