---
base_model:
- TheDrummer/Behemoth-123B-v1
- TheDrummer/Behemoth-123B-v1.1
- anthracite-org/magnum-v4-123b
library_name: transformers
tags:
- mergekit
- merge
license: other
license_name: mrl
inference: false
license_link: https://mistral.ai/licenses/MRL-0.1.md

---
![Not Horny Enough](BeMag-123B-v0.1.png)

# The Drummer becomes hornier (again)

A test merge to balance the coherency of Behemoth v1, Drummer's Uoh on v1.1 and Magnum's prose. Unsure if `task_arithmetic` works.

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

## Merge Details
### Merge Method

This model was merged using the [task arithmetic](https://arxiv.org/abs/2212.04089) merge method using [TheDrummer/Behemoth-123B-v1](https://huggingface.co/TheDrummer/Behemoth-123B-v1) as a base.

### Models Merged

The following models were included in the merge:
* [TheDrummer/Behemoth-123B-v1.1](https://huggingface.co/TheDrummer/Behemoth-123B-v1.1)
* [anthracite-org/magnum-v4-123b](https://huggingface.co/anthracite-org/magnum-v4-123b)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: TheDrummer/Behemoth-123B-v1
    parameters:
      weight: 0.425
  - model: TheDrummer/Behemoth-123B-v1.1
    parameters:
      weight: 0.425
  - model: anthracite-org/magnum-v4-123b
    parameters:
      weight: 0.15
merge_method: task_arithmetic
base_model: TheDrummer/Behemoth-123B-v1
dtype: bfloat16
```