README.md

---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: meta
    struct:
    - name: pile_set_name
      dtype: string
  splits:
  - name: train
    num_bytes: 5315790
    num_examples: 400
  download_size: 2537440
  dataset_size: 5315790
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

First 100 rows of each of `timaeus/pile-github`, `timaeus/pile-wikipedia_en`, `timaeus/pile-arxiv`, `timaeus/pile-pile-cc` in that order.