Scandinavian NER fine-tune
According to the model card, the model is a fine-tuned version of NbAiLab/nb-bert-base covering Danish, Norwegian (Bokmål and Nynorsk), Swedish, Icelandic, and Faroese.
Open Source Model Profile · saattrupdan
nbailab-base-ner-scandi
nbailab-base-ner-scandi is a 177M-parameter BERT token-classification fine-tune from saattrupdan for Scandinavian named-entity recognition.
- Publisher
- saattrupdan
- Task
- token-classification
- Model type
- bert
- License
- mit
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
nbailab-base-ner-scandi is published by saattrupdan as a bert token-classification fine-tune. The captured configuration identifies BertForTokenClassification and Safetensors metadata reports 177,270,281 parameters. According to the model card, it fine-tunes NbAiLab/nb-bert-base for named-entity recognition across Danish, Norwegian, Swedish, Icelandic, and Faroese, and card data records mit.
Recorded capabilities
Four-tag entity scheme
According to the model card, the model predicts PER, LOC, ORG, and MISC tags with publisher-supplied examples for each.
Documented training record
According to the model card, training used a learning rate of 2e-05 with batch size 8, gradient accumulation 4, and a reported 676 MB model size at about 4.16 samples per second.
Transformers token-classification setup
The hub pipeline tag is token-classification with Transformers library support, usable through the documented pipeline workflow.
Use cases in the source record
- Scandinavian named-entity extraction over Danish, Norwegian, Swedish, Icelandic, and Faroese text using the documented four-tag scheme.
- Transformers pipeline NER experiments following the model card's documented quick-start workflow.
Limitations and unknowns
- No evaluation results were extracted from this record as structured benchmarks.
- No context-window value was extracted from this record.
- Provider state is historical snapshot data and should be refreshed before being presented as current.
- Accuracy, speed, and training figures come from the publisher model card and were not independently verified by Ethen.
Source and provenance
Source: saattrupdan/nbailab-base-ner-scandi
Captured: Unknown. Processed: 2026-09-07T19:34:57.137660+00:00.
ScandiNER - Named Entity Recognition model for Scandinavian Languages Check out a demo of the model here . This model is a fine-tuned version of NbAiLab/nb-bert-base for Named Entity Recognition for Danish, Norwegian (both Bokmål and Nynorsk), Swedish, Icelandic and Faroese. It has been fine-tuned on the concatenation of DaNE , NorNE , SUC 3.0 and the Icelandic and Faroese parts of the WikiANN dataset. It also works reasonably well on English sentences, given the fact that the pretrained model is also trained on English data along with Scandinavian languages. The model will predict the following four entities: Tag Name Description P…
F001F002F003F004F005F006F007F008F009F010F011F012F013F015F016