danish-foundation-models/norwegian-dynaword

🧨 Norwegian Dynaword

Version	0.0.17 (Changelog)
Language	Norwegian (no, nor), including Bokmål (nb, nob) and Nynorsk (nn, nno)
License	Openly Licensed, See the respective dataset
Models	Currently there is no models trained on this dataset
Contact	If you have question about this project please create an issue here

Dataset Description

• Number of samples: 4.47M
• Number of tokens (Llama 3): 9.98B
• Average document length in tokens (min, max): 2.23K (2, 3.21M)

Dataset Summary

The Norwegian dynaword is a collection of Norwegian free-form text datasets from various domains. All of the datasets in the Norwegian Dynaword are openly licensed and deemed permissible for training large language models.

Norwegian dynaword is continually developed, which means that the dataset will actively be updated as new datasets become available. If you would like to contribute a dataset see the contribute section.

Loading the dataset

from datasets import load_dataset

name = "danish-foundation-models/norwegian-dynaword"
ds = load_dataset(name, split = "train")
sample = ds[1] # see "Data Instances" below

or load it by streaming the data

ds = load_dataset(name, split = "train", streaming=True)
dataset_iter = iter(ds)
sample = next(iter(dataset_iter))

You can also load a single subset at a time:

ds = load_dataset(name, "maalfrid", split = "train")

As Norwegian dynaword is continually expanding and curated you can make sure that you get the same dataset every time by specifying the revision: You can also load a single subset at a time:

ds = load_dataset(name, revision="{desired revision}")

Languages

This dataset includes the following languages:

• Norwegian (nor-Latn), including Bokmål (nob-Latn), and Nynorsk (nno-Latn)

In addition it likely contains small amounts of English due to code-switching and Danish due to the historical relation between the two languages and language misclassificaitons due to their similarity.

Language is denoted using BCP-47, using the langauge code ISO 639-3 and the script code ISO 15924.

Domains

This dynaword consist of data from various domains (e.g., legal, books, social media). The following table and figure give an overview of the relative distributions of these domains. To see a full overview of the source check out the source data section

Domain	Sources	N. Tokens
Books	ncc-books, gutenberg, wikibooks, bokselskap, nbdigital	3.42B
Spoken	stortingsforhandlingerne	2.85B
Web	maalfrid	2.23B
Report	government-nob, government-nno, public-reports	510.54M
Legal	lovdata, cellar, veidemann-municipalities, lovdata-ncc-norgeslover, lovdata-ncc-sentrale-forskrifter, lovdata-ncc-lokale-forskrifter, lovdata-ncc-odelsting, lovdata-ncc-somb-rundskriv, lovdata-ncc-rtv-rundskriv, lovdata-ncc-skatt-rundskriv, lovdata-ncc-rundskriv-lovavdeling	481.71M
Encyclopedic	wikipedia-nno, wikipedia-nob, wiki-comments, wikisource	340.46M
News	ncc-newspapers	143.73M
Total		9.98B

Language

This dynaword consist of data from various language, including Norwegian Bokmål (nob), nynorsk (nno) and Norwegian that is either mixed or where it is unknown if it is Nynorsk or Bokmål, for these we use the macrolanguage tag for Norwegian (nor) along with the individual language ids (nob, nno). The following table and figure give an overview of the relative distributions of these languages. To see a full overview of the source check out the source data section

Language	Sources	N. Tokens
nob, nno, nor	maalfrid, ncc-newspapers, stortingsforhandlingerne, wikisource	5.22B
nob	wikipedia-nob, ncc-books, government-nob	2.33B
nob, nno	cellar, bokselskap, nbdigital, veidemann-municipalities	1.96B
nob, nor, nno	public-reports, lovdata, lovdata-ncc-norgeslover, lovdata-ncc-sentrale-forskrifter, lovdata-ncc-lokale-forskrifter, lovdata-ncc-odelsting, lovdata-ncc-somb-rundskriv, lovdata-ncc-rtv-rundskriv, lovdata-ncc-skatt-rundskriv, lovdata-ncc-rundskriv-lovavdeling	325.47M
nno	wikipedia-nno, government-nno	102.64M
nno, nob	wikibooks, wiki-comments	32.98M
nno, nor, nob	gutenberg	1.55M
Total		9.98B

Licensing

The following gives an overview of the licensing in the Dynaword. To get the exact license of the individual datasets check out the overview table. These license is applied to the constituent data, i.e., the text. The collection of datasets (metadata, quality control, etc.) is licensed under CC-0.

License	Sources	N. Tokens
Other (Attribution required)	maalfrid, stortingsforhandlingerne, government-nob, government-nno, public-reports, lovdata, gutenberg, lovdata-ncc-norgeslover, lovdata-ncc-sentrale-forskrifter, lovdata-ncc-lokale-forskrifter, lovdata-ncc-odelsting, lovdata-ncc-somb-rundskriv, lovdata-ncc-rtv-rundskriv, lovdata-ncc-skatt-rundskriv, lovdata-ncc-rundskriv-lovavdeling	5.75B
CC-0	ncc-newspapers, wikipedia-nno, wikipedia-nob, ncc-books, wikibooks, wiki-comments, bokselskap, nbdigital, veidemann-municipalities	4.22B
CC-BY-SA 4.0	cellar, wikisource	3.37M
Total		9.98B

Dataset Structure

The dataset contains text from different sources which are thoroughly defined in Source Data.

Data Instances

Each entry in the dataset consists of a single text with associated metadata

{
  "id": "wikipedia-nno-0",
  "text": "'''Fredrik Hope''' () er ein norsk målmann og felespelar frå Hyen i Gloppen. Han vart leiar for Nors[...]",
  "source": "wikipedia-nno",
  "added": "2026-01-25",
  "created": "2021-01-01, 2021-12-31",
  "token_count": 214
}

Data Fields

An entry in the dataset consists of the following fields:

• id (str): A unique identifier for each document.
• text (str): The content of the document.
• source (str): The source of the document (see Source Data).
• added (str): The date when the document was added to this collection.
• created (str): The date range when the document was originally created.
• token_count (int): The number of tokens in the sample computed using the Llama 3 tokenizer.

Data Splits

The entire corpus is provided in the train split.

Dataset Creation

Curation Rationale

These datasets were collected and curated with the intention of making openly license Norwegian data available. While this was collected with the intention of developing language models it is likely to have multiple other uses such as examining language development and differences across domains.

Annotations

This data generally contains no annotation besides the metadata attached to each sample such as what domain it belongs to.

Source Data

Below follows a brief overview of the sources in the corpus along with their individual license. To get more information about the individual dataset click the hyperlink in the table.

Overview Table (click to unfold)

You can learn more about each dataset by pressing the link in the first column.

Source	Description	Domain	N. Tokens	License
stortingsforhandlingerne	OCR'd documents from the Norwegian parliament Stortinget	Spoken	2.85B	NLOD 2.0
maalfrid	Norwegian content from Norwegian institutions websites	Web	2.23B	NLOD 2.0
ncc-books	Public Domains Norwegian books from NCC	Books	1.78B	CC-0
nbdigital	Documents from Norwegian public domain books/documents from NBdigital	Books	1.61B	CC-0
veidemann-municipalities	Documents from Norwegian municipalities scraped by the Veidemann web crawler	Legal	316.96M	CC-0
government-nob	Govermental reports written on Norwegian Bokmål	Report	305.61M	NLOD 2.0
wikipedia-nob	The Norwegian Bokmål subsection of wikipedia	Encyclopedic	247.69M	CC-0
public-reports	Public reports form the NLN portal	Report	162.78M	NLOD 2.0
ncc-newspapers	OCR'd Newspapers released by the National Library of Norway (NLN)	News	143.73M	CC-0
lovdata-ncc-odelsting	Legislative documents from the Odelsting from Lovdata's CD/DVD collection, via the NCC	Legal	81.71M	NLOD 2.0
wikipedia-nno	The Norwegian Nynorsk subsection of wikipedia	Encyclopedic	60.50M	CC-0
government-nno	Govermental reports written on Norwegian Nynorsk	Report	42.14M	NLOD 2.0
lovdata	Current Norwegian laws and central regulations from Lovdata's public-data API, via the Lovverk corpus	Legal	41.23M	NLOD 2.0
bokselskap	Documents from Norwegian public domain books scraped from bokselskap.no project	Books	32.79M	CC-0
wiki-comments	Text from the comments sections of the Norwegian Wikipedia	Encyclopedic	30.96M	CC-0
lovdata-ncc-sentrale-forskrifter	Norwegian central regulations (sentrale forskrifter) from Lovdata's CD/DVD collection, via the NCC	Legal	12.03M	NLOD 2.0
lovdata-ncc-somb-rundskriv	Statements from the Norwegian Parliamentary Ombudsman (Sivilombudsmannen) from Lovdata's CD/DVD collection, via the NCC	Legal	11.56M	NLOD 2.0
lovdata-ncc-lokale-forskrifter	Norwegian local regulations (lokale forskrifter) from Lovdata's CD/DVD collection, via the NCC	Legal	5.28M	NLOD 2.0
lovdata-ncc-norgeslover	Norwegian acts of parliament (Norges Lover) from Lovdata's CD/DVD collection, via the NCC	Legal	4.02M	NLOD 2.0
lovdata-ncc-rtv-rundskriv	Circulars from the Norwegian National Insurance Administration (Rikstrygdeverket) from Lovdata's CD/DVD collection, via the NCC	Legal	3.18M	NLOD 2.0
lovdata-ncc-skatt-rundskriv	Circulars from the Norwegian Tax Administration (Skatteetaten) from Lovdata's CD/DVD collection, via the NCC	Legal	2.63M	NLOD 2.0
cellar	The official digital repository for European Union legal documents and open data	Legal	2.06M	CC-BY-SA 4.0
wikibooks	The Danish Subsection of Wikibooks	Books	2.01M	CC-0
gutenberg	The Norwegian subsection from Project Gutenberg	Books	1.55M	Gutenberg
wikisource	The Norwegian subsection of Wikisource (Wikikilden)	Encyclopedic	1.31M	CC-BY-SA 4.0
lovdata-ncc-rundskriv-lovavdeling	Circulars and statements from the Legislation Department (Lovavdelingen) of the Norwegian Ministry of Justice from Lovdata's CD/DVD collection, via the NCC	Legal	1.06M	NLOD 2.0
Total			9.98B

Data Collection and Processing

Norwegian dynaword is continually developed, which means that the dataset will actively be updated as new datasets become available. This means that the size of Dynaword increases over time as seen in the following plot:

The data collection and processing varies depending on the dataset and is documentationed the individual datasheets, which is linked in the above table. If possible the collection is documented both in the datasheet and in the reproducible script (data/{dataset}/create.py).

In addition to data specific processing we also run a series automated quality checks to ensure formatting (e.g. ensuring correctly formatted columns and unique IDs), quality checks (e.g. duplicate and empty string detection) and datasheet documentation checks. These checks are there to ensure a high quality of documentation and a minimal level of quality. To allow for the development of novel cleaning methodologies we do not provide more extensive cleaning.

Dataset Statistics

The following plot(s) are intended to give an overview of docuements length in the various sources.

Contributing to the dataset

We welcome contributions to the dataset, including new sources, improved data filtering, and other enhancements. To get started on contributing, please see the contribution guidelines

Citation Information

If you use this work, please cite the scientific article introducing the Dynaword approach and with the NCC which provides large parts of the datasets:

Enevoldsen, K.C., Jensen, K.N., Kostkan, J., Szab'o, B.I., Kardos, M., Vad, K., Heinsen, J., N'unez, A.B., Barmina, G., Nielsen, J., Larsen, R., Vahlstrup, P.B., Dalum, P.M., Elliott, D., Galke, L., Schneider-Kamp, P., & Nielbo, K.L. (2025). Dynaword: From One-shot to Continuously Developed Datasets.

Per Kummervold, Freddy Wetjen, and Javier de la Rosa. 2022. The Norwegian Colossal Corpus: A Text Corpus for Training Large Norwegian Language Models. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 3852–3860, Marseille, France. European Language Resources Association.

@article{enevoldsen2025dynaword,
  title={Dynaword: From One-shot to Continuously Developed Datasets},
  author={Enevoldsen, Kenneth and Jensen, Kristian N{\o}rgaard and Kostkan, Jan and Szab{\'o}, Bal{\'a}zs and Kardos, M{\'a}rton and Vad, Kirten and N{\'u}{\~n}ez, Andrea Blasi and Barmina, Gianluca and Nielsen, Jacob and Larsen, Rasmus and others},
  journal={arXiv preprint arXiv:2508.02271},
  year={2025}
}
@inproceedings{kummervold-etal-2022-norwegian,
    title = "The {N}orwegian Colossal Corpus: A Text Corpus for Training Large {N}orwegian Language Models",
    author = "Kummervold, Per  and
      Wetjen, Freddy  and
      de la Rosa, Javier",
    booktitle = "Proceedings of the Thirteenth Language Resources and Evaluation Conference",
    month = jun,
    year = "2022",
    address = "Marseille, France",
    publisher = "European Language Resources Association",
    url = "https://aclanthology.org/2022.lrec-1.410/",
}

Additionally, we recommend citing the relevant source datasets as well. See the individual datasheets for more information.

License information

The license for each constituent dataset is supplied in the Source data table. This license is applied to the constituent data, i.e., the text. The collection of datasets (metadata, quality control, etc.) is licensed under CC-0.

Personal and Sensitive Information

As far as we are aware the dataset does not contain information identifying sexual orientation, political beliefs, religion, or health connected along with a personal identifier of any non-public or non-historic figures.

Bias, Risks, and Limitations

Certain works in this collection are historical works and thus reflect the linguistic, cultural, and ideological norms of their time. As such, it includes perspectives, assumptions, and biases characteristic of the period, which may be considered offensive or exclusionary by contemporary standards.

Notice and takedown policy

We redistribute files shared with us under a license permitting such redistribution. If you have concerns about the licensing of these files, please contact us. If you consider that the data contains material that infringe your copyright, please:

• Clearly identify yourself with detailed contact information such as an address, a telephone number, or an email address at which you can be contacted.
• Clearly reference the original work claimed to be infringed
• Clearly identify the material claimed to be infringing and information reasonably sufficient to allow us to locate the material. You can contact us through this channel. We will comply with legitimate requests by removing the affected sources from the next release of the corpus

---

A Danish Foundation Models dataset