---
base_model:
- cs-552-2026-databand/multilingual_model
- cs-552-2026-databand/safety_model
- cs-552-2026-databand/math_model
- Qwen/Qwen3-1.7B
- cs-552-2026-databand/general_knowledge_model
library_name: transformers
tags:
- mergekit
- merge

---
# group_final

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 [Qwen/Qwen3-1.7B](https://huggingface.co/Qwen/Qwen3-1.7B) as a base.

### Models Merged

The following models were included in the merge:
* [cs-552-2026-databand/multilingual_model](https://huggingface.co/cs-552-2026-databand/multilingual_model)
* [cs-552-2026-databand/safety_model](https://huggingface.co/cs-552-2026-databand/safety_model)
* [cs-552-2026-databand/math_model](https://huggingface.co/cs-552-2026-databand/math_model)
* [cs-552-2026-databand/general_knowledge_model](https://huggingface.co/cs-552-2026-databand/general_knowledge_model)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
base_model: Qwen/Qwen3-1.7B
dtype: bfloat16
merge_method: dare_ties
modules:
  default:
    slices:
    - sources:
      - layer_range: [0, 28]
        model: cs-552-2026-databand/math_model
        parameters:
          density: 0.9
          weight: 1.0
      - layer_range: [0, 28]
        model: cs-552-2026-databand/general_knowledge_model
        parameters:
          density: 0.9
          weight: 1.0
      - layer_range: [0, 28]
        model: cs-552-2026-databand/safety_model
        parameters:
          density: 0.9
          weight: 1.0
      - layer_range: [0, 28]
        model: cs-552-2026-databand/multilingual_model
        parameters:
          density: 0.9
          weight: 1.0
      - layer_range: [0, 28]
        model: Qwen/Qwen3-1.7B
parameters:
  int8_mask: 1.0
  normalize: 1.0
```