---
license: mit
task_categories:
- text-generation
language:
- en
tags:
- bagels
pretty_name: Bagels Corpus
size_categories:
- 10M<n<100M
---

# Bagels Corpus v1

This was the dataset mix that was going to be used to train **Bagels-30M** (which does not exist yet), but after some internal testing, we found it suboptimal.

But, if you still want to download it, clone this repo:
```
git clone https://huggingface.co/datasets/fromziro/bagels_corpus
cd bagels_corpus_v1
```

Next, install the required dependencies using pip:

```bash
pip install datasets tqdm zstandard
```

Then run the download script:

```bash
python bagels_corpus.py
```

## Stats

| Dataset | Size | Share | License |
|:--|--:|--:|:--|
| [FineWeb-Edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) | 25.0 GB | 34.6% | ODC-By 1.0 |
| [DCLM Baseline 1.0](https://huggingface.co/datasets/mlfoundations/dclm-baseline-1.0) | 15.0 GB | 20.7% | CC BY 4.0 |
| [MGA FineWeb-Edu](https://huggingface.co/datasets/ByteDance-Seed/mga-fineweb-edu) | 4.0 GB | 5.5% | ODC-By 1.0 |
| [Stack v2 Edu Filtered](https://huggingface.co/datasets/common-pile/stackv2_edu_filtered) | 4.0 GB | 5.5% | Mixed |
| [peS2o Filtered](https://huggingface.co/datasets/common-pile/peS2o_filtered) | 4.0 GB | 5.5% | Mixed |
| [DOAB Filtered](https://huggingface.co/datasets/common-pile/doab_filtered) | 4.0 GB | 5.5% | CC BY / CC BY-SA |
| [PubMed Filtered](https://huggingface.co/datasets/common-pile/pubmed_filtered) | 2.0 GB | 2.8% | CC BY / CC BY-SA / CC0 |
| [arXiv Abstracts Filtered](https://huggingface.co/datasets/common-pile/arxiv_abstracts_filtered) | 2.0 GB | 2.8% | CC0 |
| [StackExchange Filtered](https://huggingface.co/datasets/common-pile/stackexchange_filtered) | 2.0 GB | 2.8% | Mixed |
| [arXiv Papers](https://huggingface.co/datasets/common-pile/arxiv_papers) | 2.0 GB | 2.8% | CC BY / CC BY-SA / CC0 |
| [SmolLM Corpus](https://huggingface.co/datasets/HuggingFaceTB/smollm-corpus) | 2.0 GB | 2.8% | ODC-By 1.0 |
| [FineMath](https://huggingface.co/datasets/HuggingFaceTB/finemath) | 2.5 GB | 3.5% | ODC-By 1.0 |
| [Wikipedia](https://huggingface.co/datasets/wikimedia/wikipedia) | 1.0 GB | 1.4% | CC BY-SA 3.0 / GFDL |
| [Project Gutenberg Filtered](https://huggingface.co/datasets/common-pile/project_gutenberg_filtered) | 1.0 GB | 1.4% | Public Domain |
| [USPTO Filtered](https://huggingface.co/datasets/common-pile/uspto_filtered) | 0.6 GB | 0.8% | Public Domain |
| [Regulations Filtered](https://huggingface.co/datasets/common-pile/regulations_filtered) | 0.6 GB | 0.8% | Public Domain |
| [USGPO Filtered](https://huggingface.co/datasets/common-pile/usgpo_filtered) | 0.6 GB | 0.8% | Public Domain |
| **Total** | **72.3 GB** | **100.0%** | |


## License

The dataset script is licensed under the MIT License and may be used freely. However, each dataset source has its own license (see above).