README.md

---
base_model:
- mlabonne/NeuralBeagle14-7B
- danish-foundation-models/munin-7b-alpha
tags:
- mergekit
- merge
license: cc-by-nc-4.0
---
# munin-neuralbeagle-7b

![](https://huggingface.co/RJuro/munin-neuralbeagle-7b/resolve/main/munin-beagle.jpg?download=true)


The model is based on danish-foundation-models/munin-7b-alpha with mlabonne/NeuralBeagle14-7B merged into using the configuration outlined below.
As per 28 January 2024, it's ranked 2nd on the [Mainland Scandinavian NLG leaderboard](https://scandeval.com/mainland-scandinavian-nlg/) (after GPT3.5)
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

## Merge Details
### Merge Method

This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [danish-foundation-models/munin-7b-alpha](https://huggingface.co/danish-foundation-models/munin-7b-alpha) as a base.

### Models Merged

The following models were included in the merge:
* [mlabonne/NeuralBeagle14-7B](https://huggingface.co/mlabonne/NeuralBeagle14-7B)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: danish-foundation-models/munin-7b-alpha
    # No parameters necessary for base model
  - model: mlabonne/NeuralBeagle14-7B
    parameters:
      density: 0.53
      weight: 0.6
merge_method: dare_ties
base_model: danish-foundation-models/munin-7b-alpha
parameters:
  int8_mask: true
dtype: bfloat16
```