---
base_model:
- ToastyPigeon/Qwen2.5-14B-Instruct-1M-Unalign
library_name: transformers
tags:
- mergekit
- merge

---
# qwen14-experimental-alt-v2

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 [TIES](https://arxiv.org/abs/2306.01708) merge method using [ToastyPigeon/Qwen2.5-14B-Instruct-1M-Unalign](https://huggingface.co/ToastyPigeon/Qwen2.5-14B-Instruct-1M-Unalign) as a base.

### Models Merged

The following models were included in the merge:
* output/tq14b-adventure-alt
* output/tq14b-fujin-alt
* output/tq14b-rp-alt
* output/tq14b-roselily-alt
* output/tq14b-gutenberg-alt

### Configuration

The following YAML configuration was used to produce this model:

```yaml
base_model: ToastyPigeon/Qwen2.5-14B-Instruct-1M-Unalign
merge_method: ties
slices:
- sources:
  - layer_range: [0, 48]
    model: output/tq14b-rp-alt
    parameters:
      density: 0.8
      weight: 0.4
  - layer_range: [0, 48]
    model: output/tq14b-roselily-alt
    parameters:
      density: 0.3
      weight: 0.2
  - layer_range: [0, 48]
    model: output/tq14b-fujin-alt
    parameters:
      density: 0.5
      weight: 0.2
  - layer_range: [0, 48]
    model: output/tq14b-gutenberg-alt
    parameters:
      density: 0.7
      weight: 0.2
  - layer_range: [0, 48]
    model: output/tq14b-adventure-alt
    parameters:
      density: 0.5
      weight: 0.1
  - layer_range: [0, 48]
    model: ToastyPigeon/Qwen2.5-14B-Instruct-1M-Unalign
```