|
--- |
|
license: cc-by-nc-sa-4.0 |
|
language: |
|
- 'no' |
|
--- |
|
|
|
Gnerative Pretrained Tranformer with 3 Billion parameters for Norwegian. |
|
|
|
It belongs to NorGLM, a suite of pretrained Norwegian Generative Language Models (NorGLMs). The model is based on GPT2 architecture. NorGLM can be used for non-commercial purposes. |
|
|
|
## Datasets |
|
|
|
All models in NorGLM are trained on 200G datasets, nearly 25B tokens, including Norwegian, Denish, Swedish, Germany and English. |
|
|
|
## Run the Model |
|
```python |
|
import torch |
|
from transformers import AutoTokenizer, AutoModelForCausalLM |
|
|
|
model_id = "NorGLM/NorGPT-3B" |
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
model = AutoModelForCausalLM.from_pretrained( |
|
model_id, |
|
device_map='auto', |
|
torch_dtype=torch.bfloat16 |
|
) |
|
|
|
text = "Tom ønsket å gå på barene med venner" |
|
inputs = tokenizer(text, return_tensors="pt") |
|
outputs = model.generate(**inputs, max_new_tokens=20) |
|
``` |
|
|
|
## Citation Information |
|
If you feel our work is helpful, please cite our paper: |
|
|
|
``` |
|
@article{liu2023nlebench+, |
|
title={NLEBench+ NorGLM: A Comprehensive Empirical Analysis and Benchmark Dataset for Generative Language Models in Norwegian}, |
|
author={Liu, Peng and Zhang, Lemei and Farup, Terje Nissen and Lauvrak, Even W and Ingvaldsen, Jon Espen and Eide, Simen and Gulla, Jon Atle and Yang, Zhirong}, |
|
journal={arXiv preprint arXiv:2312.01314}, |
|
year={2023} |
|
} |
|
``` |
