---
base_model:
- TheDrummer/UnslopSmall-22B-v1
- ReadyArt/Forgotten-Safeword-v1.1-22B
library_name: transformers
tags:
- mergekit
- merge

---
# merge

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 TIES](https://arxiv.org/abs/2311.03099) merge method using [ReadyArt/Forgotten-Safeword-v1.1-22B](https://huggingface.co/ReadyArt/Forgotten-Safeword-v1.1-22B) as a base.

### Models Merged

The following models were included in the merge:
* [TheDrummer/UnslopSmall-22B-v1](https://huggingface.co/TheDrummer/UnslopSmall-22B-v1)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
merge_method: dare_ties
base_model: ReadyArt/Forgotten-Safeword-v1.1-22B
models:
  - model: ReadyArt/Forgotten-Safeword-v1.1-22B
    parameters:
      weight: 0.7
      density: 0.8
  - model: TheDrummer/UnslopSmall-22B-v1
    parameters:
      weight: 0.3
      density: 0.4
parameters:
  int8_mask: true
  normalize: true
  lambda: 0.85
  filter:
    - pattern: "model.layers.*.self_attn"
      value: -0.15  # Direct numeric value instead of parameters dict
dtype: bfloat16
tokenizer_source: union

```