README.md

---
base_model:
- NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
- ycros/BagelMIsteryTour-v2-8x7B
- smelborp/MixtralOrochi8x7B
tags:
- mergekit
- merge

---
# maid-yuzu-v3-alter

This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).

This model was created because I wanted to know how the density and weight values of the dare_ties method affect the base model.

## 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 [smelborp/MixtralOrochi8x7B](https://huggingface.co/smelborp/MixtralOrochi8x7B) as a base.

### Models Merged

The following models were included in the merge:
* [NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO](https://huggingface.co/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO)
* [ycros/BagelMIsteryTour-v2-8x7B](https://huggingface.co/ycros/BagelMIsteryTour-v2-8x7B)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
base_model:
  model:
    path: smelborp/MixtralOrochi8x7B
dtype: bfloat16
merge_method: dare_ties
slices:
- sources:
  - layer_range: [0, 32]
    model:
      model:
        path: ycros/BagelMIsteryTour-v2-8x7B
    parameters:
      density: 0.6
      weight: 0.5
  - layer_range: [0, 32]
    model:
      model:
        path: NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO
    parameters:
      density: 0.4
      weight: 0.25
  - layer_range: [0, 32]
    model:
      model:
        path: smelborp/MixtralOrochi8x7B
```