---
dataset_info:
  features:
  - name: text
    dtype: string
  - name: is_filtered_out
    dtype: bool
  splits:
  - name: train
    num_bytes: 19005209693
    num_examples: 29451949
  download_size: 12244813118
  dataset_size: 19005209693
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---


# Dataset Card for "wikipedia-bookscorpus-en-preprocessed"

## Dataset Summary

A preprocessed and normalized combination of English Wikipedia and BookCorpus datasets, optimized for BERT pretraining. The dataset is chunked into segments of ~820 characters to accommodate typical transformer architectures.

## Dataset Details

- **Number of Examples:** 29.4 million
- **Download Size:** 12.2 GB
- **Dataset Size:** 19.0 GB

### Features:
```python
{
    'text': string,       # The preprocessed text chunk
    'is_filtered_out': bool  # Filtering flag for data quality
}
```

## Processing Pipeline

1. **Language Filtering:**
   - Retains only English language samples
   - Uses langdetect for language detection

2. **Text Chunking:**
   - Chunks of ~820 characters (targeting ~128 tokens)
   - Preserves sentence boundaries where possible
   - Splits on sentence endings (., !, ?) or spaces

3. **Normalization:**
   - Converts to lowercase
   - Removes accents and non-English characters
   - Filters out chunks < 200 characters
   - Removes special characters

4. **Data Organization:**
   - Shuffled for efficient training
   - Distributed across multiple JSONL files
   - No need for additional dataset.shuffle() during training

## Usage

```python
from datasets import load_dataset

dataset = load_dataset("shahrukhx01/wikipedia-bookscorpus-en-preprocessed")
```

## Preprocessing Details

For detailed information about the preprocessing pipeline, see the [preprocessing documentation](https://github.com/shahrukhx01/minions/tree/main/scripts/data/bert_pretraining_data/README.md).

## Limitations

- Some tokens may be lost due to chunk truncation
- Very long sentences might be split
- Some contextual information across chunk boundaries is lost

## Citation

If you use this dataset, please cite:
```
@misc{wikipedia-bookscorpus-en-preprocessed,
  author = {Shahrukh Khan},
  title = {Preprocessed Wikipedia and BookCorpus Dataset for Language Model Training},
  year = {2024},
  publisher = {GitHub & Hugging Face},
  howpublished = {\url{https://huggingface.co/datasets/shahrukhx01/wikipedia-bookscorpus-en-preprocessed}}
}
```