---
language:
- en
dataset_info:
  features:
  - name: title
    dtype: string
  - name: text
    dtype: string
  splits:
  - name: train
    num_bytes: 1183929798
    num_examples: 2110473
  download_size: 680523539
  dataset_size: 1183929798
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---

Text split version of [level 5 vital wikipedia articles](https://huggingface.co/datasets/AMead10/lvl_5_vital_wikipedia_articles). Text has been split on `\n\n`, and any instances where the length of the text was < 10 words were removed to avoid headings.