Duplicate from coral-nlp/german-commons

Co-authored-by: Lukas Gienapp <[email protected]>

.gitattributes

@@ -0,0 +1,59 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.lz4 filter=lfs diff=lfs merge=lfs -text
+*.mds filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+# Audio files - uncompressed
+*.pcm filter=lfs diff=lfs merge=lfs -text
+*.sam filter=lfs diff=lfs merge=lfs -text
+*.raw filter=lfs diff=lfs merge=lfs -text
+# Audio files - compressed
+*.aac filter=lfs diff=lfs merge=lfs -text
+*.flac filter=lfs diff=lfs merge=lfs -text
+*.mp3 filter=lfs diff=lfs merge=lfs -text
+*.ogg filter=lfs diff=lfs merge=lfs -text
+*.wav filter=lfs diff=lfs merge=lfs -text
+# Image files - uncompressed
+*.bmp filter=lfs diff=lfs merge=lfs -text
+*.gif filter=lfs diff=lfs merge=lfs -text
+*.png filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+# Image files - compressed
+*.jpg filter=lfs diff=lfs merge=lfs -text
+*.jpeg filter=lfs diff=lfs merge=lfs -text
+*.webp filter=lfs diff=lfs merge=lfs -text
+# Video files - compressed
+*.mp4 filter=lfs diff=lfs merge=lfs -text
+*.webm filter=lfs diff=lfs merge=lfs -text

DATASHEET.md

@@ -0,0 +1,262 @@
+# Datasheet: German Commons
+
+This is a datasheet compliant with the recommendations of [Gebru et al. (2018)](https://arxiv.org/abs/1803.09010v8), describing the properties of the **German Commons** dataset.
+
+## Motivation
+
+### Why was the dataset created?
+
+German Commons addresses the critical gap in large-scale open German
+text for language model training. Existing German corpora either lack
+explicit licensing, contain web-scraped content of uncertain provenance,
+or provide insufficient scale.
+
+### Has the dataset been used already?
+
+This represents the initial release of German Commons. No external usage
+has occurred prior to publication. Constituent dataset may have already been used prior.
+
+### What (other) tasks could the dataset be used for?
+
+Beyond language model pretraining, German Commons supports all German
+NLP research requiring clean, license-compliant text, multilingual model
+development, or linguistic analysis of German text across domains. The
+diverse domain coverage (legal, cultural, scientific, etc.) further
+enables domain-specific model development and cross-domain evaluation
+studies.
+
+### Who funded the creation of the dataset?
+
+Dataset compilation was supported by German and European research
+grants: German Federal Ministry of Research, Technology, and Space
+(BMFTR) under Grants  `01IS24077A`,  `01IS24077B`, and  `01IS24077D`, by
+the ScaDS.AI Center for Scalable Data Analytics and Artificial
+Intelligence, funded by the BMFTR and by the Sächsische
+Staatsministerium für Wissenschaft, Kultur und Tourismus under Grant
+ `ScaDS.AI`, and by the OpenWeb-Search.eu project, funded by the
+European Union under Grant  `GA 101070014`. Constituent datasets
+originate primarily from state-funded institutions across Germany and
+Austria.
+
+## Dataset Composition
+
+### What are the instances?
+
+Each instance represents a single German-language document with
+associated metadata and licensing information.
+
+### How many instances are there in total?
+
+The dataset contains 35,778,211 documents comprising 154,558,196,961
+GPT-2 tokens.
+
+### What data does each instance consist of?
+
+Each instance includes: a unique identifier for source
+cross-referencing, source dataset name, quality-filtered and
+paragraph-deduplicated raw text, canonical SPDX license URL, thematic
+domain key, GPT-2 token count, a perplexity score calculated using a 
+KenLM model trained on German Wikipedia text, and a OCR quality score 
+calculated using [OCRoscope](https://github.com/Pleias/OCRoscope).
+
+### Is there a label or target associated with each instance?
+
+No supervised labels exist. However, each instance contains metadata
+labels for thematic domain classification, licensing information, and
+document length statistics.
+
+### Is any information missing from individual instances?
+
+Paragraph-level deduplication may alter texts from their original form.
+Personally identifiable information has been systematically removed.
+
+### Does the dataset contain all possible instances or is it a sample (not necessarily random) of instances from a larger set?
+
+The dataset represents a filtered subset of source collections.
+Filtering removes OCR errors, extraction artifacts, and low-quality or
+duplicated content, creating a curated selection.
+
+### Are there recommended data splits?
+
+No predefined splits are provided. All data is intended for pretraining.
+
+### Are there any errors, sources of noise, or redundancies in the dataset?
+
+Despite quality filtering and deduplication, residual issues may remain:
+cross-corpus text duplicates from overlapping sources, and extraction
+artifacts from OCR and PDF-to-text processing.
+
+### Is the dataset self-contained, or does it link to or otherwise rely on external resources?
+
+The dataset is self-contained and centrally downloadable. The Source
+dataset references provided enable reproducible reconstruction.
+
+## Collection Process
+
+### What mechanisms or procedures were used to collect the data?
+
+Data collection employed multiple automated procedures: direct download
+from institutional repositories and open platforms, programmatic
+crawling via APIs where available, and automated text extraction from
+PDF and other document formats using specialized libraries. Then, the
+open source processing pipelines were applied for quality filtering and
+deduplication all sources. Validation occurred through manual inspection
+of sample outputs, cross-verification against source repositories, and
+automated consistency checks.
+
+### How was the data associated with each instance acquired?
+
+All text data represents directly observable content from original
+sources; no inference or derivation occurred. Metadata (licensing,
+thematic classification, source attribution) was extracted directly from
+source repository information or explicitly provided by institutional
+datasets. Where PDF extraction was required, raw text underwent
+validation against source documents to verify accuracy.
+
+### If the dataset is a sample from a larger set, what was the sampling strategy?
+
+Sampling was deterministic based on explicit criteria: German language
+content as per automated classification explicit open licensing, quality
+thresholds, and institutional source verification. No probabilistic
+sampling occurred; all content meeting inclusion criteria was retained
+after deduplication.
+
+### Who was involved in the data collection process and how were they compensated?
+
+Data collection was conducted by the author team using automated
+systems. No crowdworkers, contractors, or external annotators were
+employed. All processing occurred through programmatic methods without
+manual content creation or labeling requiring compensation.
+
+### Over what timeframe was the data collected? Does this timeframe match the creation timeframe of the data associated with the instances?
+
+Collection occurred between January and August 2025, using source
+dataset versions available through August 31st, 2025. The underlying
+content creation spans multiple centuries, representing a temporal range
+that significantly predates and extends beyond the collection timeframe.
+
+## Data Preprocessing
+
+### Was any preprocessing/cleaning/labeling of the data done?
+
+Comprehensive preprocessing included: text extraction from PDFs and OCR
+sources with encoding normalization, language detection and filtering
+for German content, and quality filtering targeting digitization
+artifacts and extraction errors, paragraph-level deduplication using
+content hashing, systematic PII removal, format standardization across
+all source types. Thematic domain classification was applied based on
+source dataset.
+
+### Was the raw data saved in addition to the preprocessed/cleaned/labeled data?
+
+Raw data is not provided since all constituent source datasets remain
+publicly accessible through their original repositories.
+
+### Is the software used to preprocess/clean/label the instances available?
+
+All preprocessing software is open source and available at
+<https://github.com/coral-nlp/llmdata> , ensuring complete
+reproducibility of the dataset.
+
+### Does this dataset collection/processing procedure achieve the motivation for creating the dataset stated in the first section of this datasheet?
+
+Yes. The procedure successfully addresses the identified gap by:
+providing the largest collection to-date of openly licensed German text,
+enabling open German language model development without licensing
+uncertainties, and establishing reproducible methodology for future
+dataset construction. This directly fulfills the stated motivation of
+creating license-compliant, large-scale German training data.
+
+### How will the dataset be distributed?
+
+The dataset is distributed as Parquet files through multiple public
+repositories for redundancy. Primary distribution occurs via Hugging
+Face Hub at <https://huggingface.co/datasets/coral-nlp/german-commons>.
+
+### When will the dataset be released/first distributed? What license (if any) is it distributed under?
+
+Public release occurred on 2025/10/14. Dataset metadata and compilation
+are licensed under ODC-BY 1.0 (<https://opendatacommons.org/licenses/by/1-0/>). Individual document texts retain
+their original licenses as specified in each instance's SPDX URL field,
+creating a heterogeneous but fully documented licensing structure.
+
+### Are there any copyrights on the data?
+
+Yes. Each document retains copyright under its original creator or
+institutional provider, governed by the specific license indicated in
+the instance metadata. The compilation itself does not claim additional
+copyright over constituent texts.
+
+### Are there any fees or access/export restrictions?
+
+The dataset is freely accessible without fees or registration
+requirements. However, users must comply with individual document
+licenses, which may include attribution requirements or share-alike
+provisions. Commercial use is permitted by all constituent licenses.
+
+## Dataset Maintenance
+
+### Who is supporting/hosting/maintaining the dataset?
+
+The dataset is maintained by the authors of this report.
+
+### Will the dataset be updated? If so, how often and by whom?
+
+Updates may occur when significant new German open-source collections
+become available. The original authors will coordinate updates, with
+community contributions welcomed through the open-source pipeline.
+
+### How will updates be communicated?
+
+Updates will be announced through: versioned releases on hosting
+platforms with detailed changelogs, academic publication updates when
+substantial changes occur.
+
+### If the dataset becomes obsolete how will this be communicated?
+
+Obsolescence will be communicated through deprecation notices on all
+hosting platforms.
+
+### Is there a repository to link to any/all papers/systems that use this dataset?
+
+No centralized usage repository will be maintained. Usage tracking
+occurs through standard academic citation of the dataset paper. Users
+are encouraged to cite the dataset publication when reporting results or
+building derivative works.
+
+### If others want to extend/augment/build on this dataset, is there a mechanism for them to do so?
+
+The open-source `llmdata` pipeline enables community extensions through
+standardized data ingestion protocols for new sources and automated
+quality assessment and deduplication using established filtering
+criteria. Community contributions undergo review by the maintenance
+team.
+
+## Ethical Considerations
+
+### Were any ethical review processes conducted?
+
+No formal institutional review board process was conducted. The dataset
+relies exclusively on pre-existing, publicly available, and explicitly
+licensed materials from established institutional sources. Data
+processing incorporated ethical considerations including systematic PII
+removal and exclusion of sources lacking clear licensing frameworks.
+
+### Does the dataset contain data that might be considered confidential?
+
+No. All included content derives from explicitly open-licensed
+institutional sources.
+
+### Does the dataset contain data that, if viewed directly, might be offensive, insulting, threatening, or might otherwise cause anxiety?
+
+Potentially yes. The dataset spans centuries of German text documents,
+which may include historical perspectives, political viewpoints, or
+language that could be considered offensive by contemporary standards.
+The scale and temporal range make comprehensive content moderation
+infeasible. Users should exercise appropriate caution.
+
+### Does the dataset relate to people?
+
+The dataset may contain publicly available information relating to
+individuals in various contexts including historical documents,
+biographical information, academic citations, and government records.

README.md

@@ -0,0 +1,479 @@
+---
+annotations_creators:
+- machine-generated
+language_creators:
+- found
+language:
+- de
+license:
+- odc-by
+multilinguality:
+- monolingual
+size_categories:
+- 100B<n
+source_datasets: []
+task_categories:
+- text-generation
+task_ids:
+- language-modeling
+pretty_name: German Commons
+tags:
+- german
+- commons
+- legal
+- scientific
+- cultural
+- political
+- web
+- news
+- economic
+configs:
+- config_name: default
+  data_files: subset=*/source=*/*.parquet
+  default: true
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+  splits:
+  - name: train
+    num_examples: 36773579
+- config_name: cultural
+  data_files:
+  - split: blbooks
+    path: subset=Cultural/source=BLBooks/*.parquet
+  - split: dibilit
+    path: subset=Cultural/source=DiBiLit/*.parquet
+  - split: dibiphil
+    path: subset=Cultural/source=DiBiPhil/*.parquet
+  - split: germanpd
+    path: subset=Cultural/source=GermanPD/*.parquet
+  - split: mosel
+    path: subset=Cultural/source=MOSEL/*.parquet
+  - split: sbbfulltexts
+    path: subset=Cultural/source=SBB Fulltexts/*.parquet
+  - split: wikisource
+    path: subset=Cultural/source=Wikisource/*.parquet
+  - split: wikivoyage
+    path: subset=Cultural/source=Wikivoyage/*.parquet
+  - split: wikiquote
+    path: subset=Cultural/source=Wikiquote/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: economic
+  data_files:
+  - split: tedeutenders
+    path: subset=Economic/source=TEDEUTenders/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: legal
+  data_files:
+  - split: bverfgaes
+    path: subset=Legal/source=Amtliche Entscheidungssammlung des Bundesverfassungsgerichts/*.parquet
+  - split: bundesrecht
+    path: subset=Legal/source=Deutsches Bundesrecht/*.parquet
+  - split: bag
+    path: subset=Legal/source=Entscheidungen des Bundesarbeitsgerichts/*.parquet
+  - split: bfh
+    path: subset=Legal/source=Entscheidungen des Bundesfinanzhofs/*.parquet
+  - split: bgh
+    path: subset=Legal/source=Entscheidungen des Bundesgerichtshofs/*.parquet
+  - split: bgh20
+    path: subset=Legal/source=Entscheidungen des Bundesgerichtshofs in Strafsachen
+      aus dem 20. Jahrhundert/*.parquet
+  - split: bpatg
+    path: subset=Legal/source=Entscheidungen des Bundespatentgerichts/*.parquet
+  - split: bverfg
+    path: subset=Legal/source=Entscheidungen des Bundesverfassungsgerichts/*.parquet
+  - split: bverwg
+    path: subset=Legal/source=Entscheidungen des Bundesverwaltungsgerichts/*.parquet
+  - split: eurlex
+    path: subset=Legal/source=EurLEX/*.parquet
+  - split: openlegaldata
+    path: subset=Legal/source=Open Legal Data/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: news
+  data_files:
+  - split: anno
+    path: news_anno_*.parquet
+  - split: zeitungsportal
+    path: news_deutsches_zeitungsportal_*.parquet
+  - split: europeana
+    path: news_europeana_newspapers_*.parquet
+  - split: wikinews
+    path: news_wikinews_*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: political
+  data_files:
+  - split: btdrucksachen
+    path: subset=Political/source=Drucksachen des Bundestages/*.parquet
+  - split: eurovoc
+    path: subset=Political/source=EuroVoc/*.parquet
+  - split: germanpoliticalspeeches
+    path: subset=Political/source=German Political Speeches/*.parquet
+  - split: btplenarprotokolle
+    path: subset=Political/source=Plenarprotokolle des Bundestages/*.parquet
+  - split: reichtagsprotokolle
+    path: subset=Political/source=Reichtagsprotokolle/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: scientific
+  data_files:
+  - split: arxiv
+    path: subset=Scientific/source=arXiv/*.parquet
+  - split: doab
+    path: subset=Scientific/source=Directory of Open Access Books/*.parquet
+  - split: polyjournal
+    path: subset=Scientific/source=Polytechnisches Journal/*.parquet
+  - split: wikibooks
+    path: subset=Scientific/source=Wikibooks/*.parquet
+  - split: openalex
+    path: subset=Scientific/source=OpenAlex/*.parquet
+  - split: wikiversity
+    path: subset=Scientific/source=Wikiversity/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+- config_name: web
+  data_files:
+  - split: onemillionposts
+    path: subset=Web/source=One Million Posts/*.parquet
+  - split: thestack
+    path: subset=Web/source=The Stack/*.parquet
+  - split: wikidiscussions
+    path: subset=Web/source=Wiki Discussions/*.parquet
+  - split: wikipedia
+    path: subset=Web/source=Wikipedia/*.parquet
+  - split: youtubecommons
+    path: subset=Web/source=Youtube Commons/*.parquet
+  features:
+  - name: id
+    dtype: string
+  - name: source
+    dtype: string
+  - name: subset
+    dtype:
+      class_label:
+        names:
+        - Cultural
+        - Economic
+        - Legal
+        - News
+        - Political
+        - Scientific
+        - Web
+  - name: text
+    dtype: string
+  - name: license
+    dtype:
+      sequence: string
+  - name: num_tokens
+    dtype: int64
+  - name: perplexity
+    dtype: float64
+  - name: ocr_score
+    dtype: int64
+---
+
+# German Commons - 154 Billion Tokens of Openly Licensed Text for German Language Models
+
+A comprehensive collection of German-language text data under open licenses for training German language models.
+
+- **Datasheet**: [DATASHEET.md](DATASHEET.md).
+- **Paper**: [arxiv.org/abs/2510.13996](https://arxiv.org/abs/2510.13996)
+- **Code**: [github.com/coral-nlp/llmdata](https://github.com/coral-nlp/llmdata)
+- **Bloom Filter** (DOLMA-compatible): [bloom_filter.bin](bloom_filter.bin)
+
+## Dataset Description
+
+This dataset is aggregated from **41 diverse sources** and contains **154.56 billion tokens** of German text data with **35.78 million documents** spanning **7 thematic domains**:
+
+- 🌐 **Web Commons**: 19.89B tokens source from Wiki projects, online discussions, code repositories, social media posts, YouTube transcripts
+- 💬 **Political Commons**: 3.57B tokens sourced from parliamentary documents, speeches, protocols, political vocabulary
+- ⚖️ **Legal Commons**: 2.99B tokens sourced from court decisions, federal law, legal databases, EU legal documents
+- 📰 **News Commons**: 72.67B tokens sourced from historical and current newspapers archives
+- 🏦 **Economics Commons**: 0.11B tokens sourced from EU public tenders
+- 📚 **Cultural Commons**: 54.49B tokens sourced from cultural heritage collections
+- 🔬 **Scientific Commons**: 0.84B tokens sourced from scholarly papers, books, and technical journals
+
+## Dataset Features
+
+Each record contains the following fields:
+
+- **id**: Unique identifier string, as per each documents' source dataset
+- **source**: Source dataset name
+- **subset**: Thematic subset (Cultural, Legal, Political, Scientific, News, Web, Economic)
+- **text**: Main text content; deduplicated, quality filtered, with consistent formatting and encoding. Can be split at newlines to obtain paragraph text.
+- **license**: List of applicable licenses for each document, given as canonical SPDX license URL.
+- **num_tokens**: GPT-2 token count
+- **perplexity**: Text perplexity measured with a KenLM model trained on German Wikipedia text
+- **ocr_score**: OCR quality score measured using [OCRoscope](https://github.com/Pleias/OCRoscope)
+
+## Dataset Usage
+
+- Load the entire dataset
+
+  ```python
+  from datasets import load_dataset
+  
+  ds = load_dataset("coral-nlp/german-commons")
+  ```
+
+- Load a thematic subset
+
+  ```python
+  ds = load_dataset("coral-nlp/german-commons", "cultural")
+  ```
+
+- Load individual source datasets
+
+  ```python
+  wikipedia = load_dataset("coral-nlp/german-commons", "web", split="wikipedia")
+  ```
+
+Supported splits and constituent datasets are:
+
+| Subset | Split Key | Dataset Name | Docs | Tokens | License | Text Type | Source |
+|:-------------|:----------------------------|:------------------------------------------------------|----------:|---------------:|:-------------|:---------------------------|:----------------------------------------------------------------- |
+| `web` | `wikipedia` | Wikipedia | 2,930,224 | 2,948,751,608 | CC-BY-SA-4.0 | Various | [🔗](https://zenodo.org/records/14748605) |
+| `web` | `wikidiscussions` | Wikipedia Discussions | 8,349,076 | 1,218,210,917 | CC-BY-SA-4.0 | Online Discussions | [🔗](https://corpora.ids-mannheim.de/pub/wikipedia-deutsch/2024/) |
+| `web` | `youtubecommons` | YouTube Commons | 2,809,714 | 14,478,850,964 | Various | Video Subtitles | [🔗](https://huggingface.co/datasets/PleIAs/YouTube-Commons) |
+| `web` | `onemillionposts` | One Million Posts Corpus | 946,082 | 94,872,633 | CC-BY-4.0 | Online Discussions | [🔗](https://ofai.github.io/million-post-corpus/) |
+| `web` | `thestack` | The Stack (Markdown and TXT Subsets) | 421,466 | 1,105,173,228 | Various | Various | [🔗](https://huggingface.co/datasets/bigcode/the-stack-dedup) |
+| `political` | `reichtagsprotokolle` | Reichtagsprotokolle | 522 | 703,495,637 | CC-BY-SA-4.0 | Parliamentary Protocols | [🔗](https://zenodo.org/records/10225467) |
+| `political` | `germanpoliticalspeeches` | German Political Speeches | 6,678 | 29,409,655 | CC-BY-4.0 | Speech Transcripts | [🔗](https://zenodo.org/records/3611246) |
+| `political` | `btdrucksachen` | Corpus der Drucksachen des Deutschen Bundestages | 3,017 | 528,769,669 | CC0-1.0 | Parliamentary Publications | [🔗](https://zenodo.org/records/4643066) |
+| `political` | `btplenarprotokolle` | Corpus der Plenarprotokolle des Deutschen Bundestages | 1,833 | 316,034,708 | CC0-1.0 | Parliamentary Protocols | [🔗](https://zenodo.org/records/4542662) |
+| `political` | `eurovoc` | EuroVoc | 245,838 | 1,988,111,462 | EUPL | Parliamentary Publications | [🔗](https://huggingface.co/datasets/EuropeanParliament/Eurovoc) |
+| `legal` | `bundesrecht` | Corpus des Deutschen Bundesrechts | 3,217 | 1,004,294 | CC0-1.0 | German Federal Laws | [🔗](https://zenodo.org/records/14592346) |
+| `legal` | `openlegaldata` | OpenLegalData | 249,909 | 1,915,956,613 | CC0-1.0 | Court Decisions | [🔗](https://huggingface.co/datasets/schneiderai/openlegaldata) |
+| `legal` | `bfh` | Corpus der Entscheidungen des BFH | 10,885 | 67,791,931 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/14622341) |
+| `legal` | `bgh20` | Entscheidungen des BGH in Strafsachen des 20. Jhd. | 36,062 | 92,873,390 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/4540377) |
+| `legal` | `bgh` | Corpus der Entscheidungen des BGH | 77,258 | 292,832,709 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/12814022) |
+| `legal` | `bverfg` | Corpus der Entscheidungen des BVerfG | 8,028 | 39,503,223 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/12705674) |
+| `legal` | `bpatg` | Corpus der Entscheidungen des BpatG | 30,705 | 185,099,188 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/10849977) |
+| `legal` | `bverwg` | Corpus der Entscheidungen des BVerwG | 27,185 | 123,487,739 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/10809039) |
+| `legal` | `bverfgaes` | Corpus der amtl. Entscheidungssammlung des BVerfG | 919 | 24,427,294 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/10783177) |
+| `legal` | `bag` | Corpus der Entscheidungen des BAG | 5,624 | 48,248,111 | CC0-1.0 | Court Decisions | [🔗](https://zenodo.org/records/4006645) |
+| `legal` | `eurlex` | EurLEX | 64,934 | 201,263,562 | CC-BY-4.0 | European Union Laws | [🔗](https://zenodo.org/record/5363165/) |
+| `news` | `zeitungsportal` | Deutsches Zeitungsportal | 8,076,164 | 43,871,094,547 | CC0-1.0 | News Articles | [🔗](https://www.deutsche-digitale-bibliothek.de/newspaper) |
+| `news` | `europeana` | Europeana Newspapers | 3,256,341 | 20,684,418,365 | CC0-1.0 | News Articles | [🔗](https://huggingface.co/datasets/biglam/europeana_newspapers) |
+| `news` | `anno` | ANNO | 1,910,281 | 8,103,825,248 | CC0-1.0 | News Articles | [🔗](https://labs.onb.ac.at/en/datasets/anno/) |
+| `news` | `wikinews` | WikiNews | 23,266 | 14,222,520 | CC-BY-4.0 | News Articles | [🔗](https://de.wikinews.org) |
+| `economic` | `tedeutenders` | TEDEUTenders | 57,214 | 110,611,112 | CC0-1.0 | Procurement Notices | [🔗](https://huggingface.co/datasets/PleIAs/TEDEUTenders) |
+| `cultural` | `dibilit` | DiBiLit-Korpus | 2,062 | 216,391,448 | CC-BY-SA-4.0 | Literature | [🔗](https://zenodo.org/records/5786725) |
+| `cultural` | `dibiphil` | DiBiPhil-Korpus | 269 | 32,151,997 | CC-BY-SA-4.0 | Literature | [🔗](https://github.com/deutschestextarchiv/DiBiPhil) |
+| `cultural` | `wikisource` | Wikisource | 240,689 | 347,770,430 | CC-BY-SA-4.0 | Various | [🔗](https://dumps.wikimedia.org/dewikisource/20250801/) |
+| `cultural` | `wikivoyage` | Wikivoyage | 20,370 | 42,025,478 | CC-BY-SA-4.0 | Travel | [🔗](https://zenodo.org/records/14748553) |
+| `cultural` | `germanpd` | German-PD | 123,592 | 49,333,198,231 | CC0-1.0 | Literature | [🔗](https://huggingface.co/datasets/PleIAs/German-PD) |
+| `cultural` | `blbooks` | BLBooks | 3,714 | 1,012,047,216 | CC0-1.0 | Literature | [🔗](https://huggingface.co/datasets/biglam/blbooks-parquet) |
+| `cultural` | `mosel` | MOSEL | 3,127,203 | 3,181,917,752 | CC-BY-4.0 | Speech Transcripts | [🔗](https://huggingface.co/datasets/FBK-MT/mosel) |
+| `cultural` | `sbbfulltexts` | SBB Fulltexts | 2,605,569 | 358,514,283 | CC-BY-4.0 | Literature | [🔗](https://zenodo.org/records/7716098) |
+| `cultural` | `wikiquote` | Wikiquote | 8,612 | 6,688,458 | CC-BY-4.0 | Quotes & Proverbs | [🔗](https://de.wikiquote.org) |
+| `scientific` | `wikibooks` | Wikibooks | 346 | 180,257,799 | CC-BY-SA-4.0 | Educational Books | [🔗](https://zenodo.org/records/14748586) |
+| `scientific` | `polyjournal` | Digitalisierung des Polytechnischen Journals | 27,292 | 50,434,996 | CC-BY-SA-4.0 | Scholarly Papers | [🔗](https://github.com/deutschestextarchiv/dingler) |
+| `scientific` | `doab` | Directory of Open Access Books | 1,939 | 166,920,321 | Various | Scholarly Books | [🔗](https://www.doabooks.org) |
+| `scientific` | `arxiv` | arXiv | 8 | 103,478 | Various | Scholarly Papers | [🔗](https://www.kaggle.com/datasets/Cornell-University/arxiv) |
+| `scientific` | `openalex` | OpenAlex | 47,733 | 413,632,648 | Various | Educational Content | [🔗](https://de.wikiversity.org) |
+| `scientific` | `wikiversity` | Wikiversity | 16,371 | 27,802,099 | CC-BY-SA-4.0 | Scholarly Papers | [🔗](https://openalex.org) |
+
+## Citation
+
+If you use this dataset, please cite the correponding paper:
+
+```bibtex
+@article{gienapp:2025d,
+	title        = {{The German Commons -- 154 Billion Tokens of Openly Licensed Text for German Language Models}},
+	author       = {Lukas Gienapp and
+                    Christopher Schr\"oder and
+                    Stefan Schweter and
+                    Christopher Akiki and
+                    Ferdinand Schlatt and
+                    Arden Zimmermann and
+                    Phillipe Gen\^et and
+                    Martin Potthast},
+	year         = 2025,
+	month        = oct,
+	journal      = {CoRR},
+	volume       = {abs/2510.13996},
+	url          = {https://arxiv.org/abs/2510.13996}
+}
+```
+
+## License
+
+This dataset aggregation and metadata is released under ODC-BY license. Individual documents have their own specific licenses - please check the `license` field for each record.

bloom_filter.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57c1b680ffef03de26ec1c2727bad141e24ad5d2904795cb1beae330831537cb
+size 4294968084

subset=Cultural/source=BLBooks/part-0001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23031935983f04f48bd66e6537357311141f99bf1ca711ed9025f8da1acd2bb6
+size 1354078008

subset=Cultural/source=BLBooks/part-0002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b427832d12504843698333889d4851b6cb5293e4addaf1c08f92757884187cd1
+size 308444014

subset=Cultural/source=DiBiLit/part-0001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:85ba3cb6d2b526c1e1c5d6b59a6e7c9a75be3039ab26e65b1c3d224fb9f6c32f
+size 342802369

subset=Cultural/source=DiBiPhil/part-0001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c32e2a51ec0721b2aac00f869101dae43666d32af268594fb03139ba71413033
+size 47453198

subset=Cultural/source=GermanPD/part-0001.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:193e7695fb8076af9ec29778f4a4c6d662abe9aaa5476608571d425b4b783f70
+size 676504492

subset=Cultural/source=GermanPD/part-0002.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d1d01d126c2d3910761f8d49a0e3a8a4c6b89f186e78b5d887b4eba536e8e97
+size 1443734821

subset=Cultural/source=GermanPD/part-0003.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7359a6e07e11cb0eb0e5c054143dc70fa23f3ce4d05e25e309045c2c4e37134b
+size 1439412386

subset=Cultural/source=GermanPD/part-0004.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a10407853185fe74f0a1e26717d05c1f66b9542c37e1aea40adf7b920819f71e
+size 1443477926

subset=Cultural/source=GermanPD/part-0005.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1c12f66f221e01b9582e6aee32e8a54d9976ee4013c1ff6ba907ef29b52b385
+size 1438692783

subset=Cultural/source=GermanPD/part-0006.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4bb67e8d04da35a266940612c4050bf76f2847ea099325fbd6beccda9dd49c6
+size 1431459189

subset=Cultural/source=GermanPD/part-0007.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b1270af8a68f44c5f8c61f618b7bf52cb2df86c44a47d2b30ad18978aed4adf9
+size 1434871205

subset=Cultural/source=GermanPD/part-0008.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87cdc2d5a4c9bdbde704040c1530462490fd67cebeb24659767c283e9808c152
+size 1418445890

subset=Cultural/source=GermanPD/part-0009.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69c927dd9e166280c86e16aa5b56289743a84795360c22afdd48800079b7cf82
+size 1436818312

subset=Cultural/source=GermanPD/part-0010.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41cd63f6c21bd79fdf569e515d708ce02e5b30c6891512346ea03b0e4859b835
+size 1455947561

subset=Cultural/source=GermanPD/part-0011.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1ce8552504260a393fba2e95ec11af17e694dc7ddcdd4cbf10bc45723bae0ff
+size 1454247208

subset=Cultural/source=GermanPD/part-0012.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a855d7b938afc0595422bcd78c003a27831ffab1ed4cad4ed5c7912dc6cf73dc
+size 1438373053

subset=Cultural/source=GermanPD/part-0013.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9614923eb917dfec8108fea2cf8d3600c2c4418b377aa977f056d21b354690d
+size 1439958630

subset=Cultural/source=GermanPD/part-0014.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8257a9eb49cd640b3a8980c83caa79c43a3ccaf835a41d1be182b4ac6a573878
+size 1452069614

subset=Cultural/source=GermanPD/part-0015.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:043585507fd18e08c07f8759d133d5ee19c541dbe7eff4de1fffc0dce6264e42
+size 1446016474

subset=Cultural/source=GermanPD/part-0016.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2438ac580fcfe5e60fe4bf9a8e0a4e30fec9b1b485a4684504196361716b02e7
+size 1432128126

subset=Cultural/source=GermanPD/part-0017.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d5acd8226e5a1c637d6d78bd38636683b135fe45b0996326fa22c410b67f780
+size 1426189384

subset=Cultural/source=GermanPD/part-0018.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:112709779b82c2ec0342019f2edbb642fccfb1ff24e9c2db8c059e88b0e5e785
+size 1440983089

subset=Cultural/source=GermanPD/part-0019.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af7f77e2399d49692bfeee2a776b7884be9ee867dca15b8f8131b2f5d7024ecb
+size 933083353

subset=Cultural/source=GermanPD/part-0020.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:686240f2053cf799a5659d19037ed65f8dc96ec2805445d7769d2e5ba60085c1
+size 1438203248

subset=Cultural/source=GermanPD/part-0021.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7216e6b659100be253e376714683acdd4ef17a49e354e802ea3f2d9264c10926
+size 1455350364

subset=Cultural/source=GermanPD/part-0022.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aefb01e9deb5df7c902a81146c9afe79619a493aa41910dddd5006dc07472b97
+size 1458100415

subset=Cultural/source=GermanPD/part-0023.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c33100130b1eb952d2424bbab511a97729f54da80e49329931a1345112cbe31c
+size 1446591112

subset=Cultural/source=GermanPD/part-0024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8f8d056cce322f53aa475317dd97edac784e506eba6b9fcab0d403d54ecb193
+size 1429167089

subset=Cultural/source=GermanPD/part-0025.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae24efdb65c05e38302852ecf247881a995caa5409de0dec1a677497a3ed9467
+size 1428744082

subset=Cultural/source=GermanPD/part-0026.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00c32c1113b3164301112e776b75c08f84f54dc6d2185ce00ae0bfd5a2499086
+size 1447087708

subset=Cultural/source=GermanPD/part-0027.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:674fcfd67947ed8c812e80b5f20f65fa541fb8b2952190d7b2505796e0fb182c
+size 1433796706

subset=Cultural/source=GermanPD/part-0028.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16846c5948689bc5bfc091e33aec1c7582b88dc8a88af400adc2143f43195787
+size 1443763802

subset=Cultural/source=GermanPD/part-0029.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:053bcb1f87f727651aaf3f1c4cde8632270585ad722cdace0b73bd6858153847
+size 1442798042

subset=Cultural/source=GermanPD/part-0030.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d60b49e6cc47d708425f0017ae59a648588c5e531b3524dda41ad246f1fb2699
+size 1430323209

subset=Cultural/source=GermanPD/part-0031.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:814690adf5f2f6d15707c72330dd1cd9e9738802e23f7e5a97d844488859f767
+size 1450730905

subset=Cultural/source=GermanPD/part-0032.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b967ac7820aed56c01ecdddbd63a7dce36091ab20e51dbc144dc8ed0881db50
+size 1456149536

subset=Cultural/source=GermanPD/part-0033.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3cb7d35a4f6e67cfe54f3a5764742d99e5a10d4ac1c377cd54490bbb8643b0d
+size 1437512772

subset=Cultural/source=GermanPD/part-0034.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05cec5af5e7745943425e924d3731b45ecfe00f97e38c32a3acdbf657bea4b4b
+size 1441439359

subset=Cultural/source=GermanPD/part-0035.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d3bed83330cc1a0eef822a5f647f9da556bae58f66509bc35f1737a9d69b0417
+size 1449255467

subset=Cultural/source=GermanPD/part-0036.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34748d66f09435ce0554d7f97ebbe533a96bb9bb2e3617f39b48a5d82eedfc25
+size 1448007954

subset=Cultural/source=GermanPD/part-0037.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22d8dd522ce49049fc33c19052ca79309b8aaad229b863fc68cc48c78c7a75b8
+size 1439027562

subset=Cultural/source=GermanPD/part-0038.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9beda007237a1b46b38cc46e9ed47fb0d3afe6dda89c6bed9202aa4dc341fc44
+size 1444127810

subset=Cultural/source=GermanPD/part-0039.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39283c70e69b4d169549fbee97cb261493497579c0bd70644384dad3f65c350b
+size 1440732996

subset=Cultural/source=GermanPD/part-0040.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea4dc06ed7dc66db95f8208f7fe4b7d902a9f2fdce684afaa481a3dde7bed4d2
+size 1445673466

subset=Cultural/source=GermanPD/part-0041.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2726201365021930c57074993b06bd17e9d19f805ab1f66faee1745b70bb76e1
+size 1438043290

subset=Cultural/source=GermanPD/part-0042.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:982a26f03b3391343bf579b5ac734dd78d5348940ef1398875e905c971c01bc1
+size 1450981707