FinGPT Large
FinGPT Large is a Finnish text-generation base model in TurkuNLP's Finnish GPT-3 series. It uses the BLOOM architecture and is not an OpenAI GPT-3 checkpoint.
Model design and training
24 layers, width 1,536 and 16 attention heads. The card reports 300 billion training tokens. The Finnish corpus combines web text, news, books, Wikipedia and discussion data. This is a pure language model, without dialogue or question-answering instruction tuning.gpt3-finnish-large: official model card
Practical assessment
Practical guidance. Large expands width while keeping Medium's layer count. That distinction is useful when comparing model capacity under a shared training approach. A text-completion prompt is appropriate for the base release; an assistant dialogue format alone does not provide instruction training.
Specifications & access
| Detail | Value and conditions |
|---|---|
| Access | Downloadable |
| Model identifiers | gpt3-finnish-largeFinnish GPT-3 Large |
| Model size | 881 million parameters.gpt3-finnish-large: official model card |
| License | Apache 2.0.gpt3-finnish-large: official model card |
| Training state | Pretrained base; not instruction-tuned.gpt3-finnish-large: official model card |
Related models
- FinGPT 13B
TurkuNLP · FinGPT
FinGPT 13B is a Finnish text-generation base model in TurkuNLP's Finnish GPT-3 series.
Launch date not confirmedDownloadable - FinGPT Small
TurkuNLP · FinGPT
FinGPT Small is a Finnish text-generation base model in TurkuNLP's Finnish GPT-3 series.
Launch date not confirmedDownloadable