| | --- |
| | language: no |
| | license: cc-by-4.0 |
| | pipeline_tag: fill-mask |
| | tags: |
| | - norwegian |
| | - bert |
| | thumbnail: https://raw.githubusercontent.com/ltgoslo/NorBERT/main/Norbert.png |
| | widget: |
| | - text: "Nå ønsker de seg en [MASK] bolig. " |
| |
|
| | --- |
| | ## Quickstart |
| | **Release 2.0** (February 7, 2022) |
| |
|
| | Please check also our newer models: [NorBERT 3 family](https://huggingface.co/ltg/norbert3-base), trained with a better architecture. |
| |
|
| | Trained on the very large corpus of Norwegian (C4 + NCC, about 15 billion word tokens). |
| | Features a 50 000 words vocabulary and was trained using Whole Word Masking. |
| |
|
| | Download the model here: |
| | * Cased Norwegian BERT Base 2.0 (NorBERT 2): [221.zip](http://vectors.nlpl.eu/repository/20/221.zip) |
| |
|
| | More about NorBERT training corpora, training procedure and evaluation benchmarks: http://norlm.nlpl.eu/ |
| |
|
| | Associated code: https://github.com/ltgoslo/NorBERT |
| |
|
| | Check this paper for more details: |
| |
|
| | _Andrey Kutuzov, Jeremy Barnes, Erik Velldal, Lilja Øvrelid, Stephan Oepen. [Large-Scale Contextualised Language Modelling for Norwegian](https://aclanthology.org/2021.nodalida-main.4/), NoDaLiDa'21 (2021)_ |
| |
|
| | NorBERT was trained as a part of NorLM, a joint initiative of the projects [EOSC-Nordic](https://www.eosc-nordic.eu/) (European Open Science Cloud), |
| | coordinated by the [Language Technology Group](https://www.mn.uio.no/ifi/english/research/groups/ltg/) (LTG) at the University of Oslo. |
| |
|
| | The computations were performed on resources provided by UNINETT Sigma2 - the National Infrastructure for High Performance Computing and Data Storage in Norway. |
| |
|
| | ## NorBERT-3 |
| | In 2023, we released a new family of *NorBERT-3* language models for Norwegian. In general, we now recommend using these models: |
| |
|
| | - [NorBERT 3 xs](https://huggingface.co/ltg/norbert3-xs) (15M parameters) |
| | - [NorBERT 3 small](https://huggingface.co/ltg/norbert3-small) (40M parameters) |
| | - [NorBERT 3 base](https://huggingface.co/ltg/norbert3-base) (123M parameters) |
| | - [NorBERT 3 large](https://huggingface.co/ltg/norbert3-large) (323M parameters) |
| |
|
| | NorBERT-3 is described in detail in this paper: |
| | [NorBench – A Benchmark for Norwegian Language Models](https://aclanthology.org/2023.nodalida-1.61/) (Samuel et al., NoDaLiDa 2023) |
| |
|
