---
base_model:
- TareksLab/Doppleganger-V8-LLaMa-70B
- TareksLab/Dungeons-R1-LLaMa-70B
- TareksLab/Dragons-V1-LLaMa-70B
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 [SCE](https://arxiv.org/abs/2408.07990) merge method using [TareksLab/Doppleganger-V8-LLaMa-70B](https://huggingface.co/TareksLab/Doppleganger-V8-LLaMa-70B) as a base.

### Models Merged

The following models were included in the merge:
* [TareksLab/Dungeons-R1-LLaMa-70B](https://huggingface.co/TareksLab/Dungeons-R1-LLaMa-70B)
* [TareksLab/Dragons-V1-LLaMa-70B](https://huggingface.co/TareksLab/Dragons-V1-LLaMa-70B)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
models:
  - model: TareksLab/Dungeons-R1-LLaMa-70B
    parameters:
      weight: 0.35
      density: 0.5
      select_topk: 0.5
      lambda: 1.0
  - model: TareksLab/Dragons-V1-LLaMa-70B
    parameters:
      weight: 0.35
      density: 0.5
      select_topk: 0.5
      lambda: 1.0
  - model: TareksLab/Doppleganger-V8-LLaMa-70B
    parameters:
      weight: 0.30
      density: 0.5
      select_topk: 0.5
      lambda: 1.0
base_model: TareksLab/Doppleganger-V8-LLaMa-70B
merge_method: sce
parameters:
  normalize: false
tokenizer:
  source: base
chat_template: llama3
dtype: bfloat16

```