---
base_model:
- rombodawg/Rombos-LLM-V2.5-Qwen-72b
- Qwen/Qwen2.5-72B
- abacusai/Dracarys2-72B-Instruct
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 [TIES](https://arxiv.org/abs/2306.01708) merge method using [Qwen/Qwen2.5-72B](https://huggingface.co/Qwen/Qwen2.5-72B) as a base.

### Models Merged

The following models were included in the merge:
* [rombodawg/Rombos-LLM-V2.5-Qwen-72b](https://huggingface.co/rombodawg/Rombos-LLM-V2.5-Qwen-72b)
* [abacusai/Dracarys2-72B-Instruct](https://huggingface.co/abacusai/Dracarys2-72B-Instruct)

### Configuration

The following YAML configuration was used to produce this model:

```yaml
# Qwen2.5-72B-2x-Instruct-TIES-v1.0

models:
  - model: abacusai/Dracarys2-72B-Instruct # Coding enhancement Fine-tuning
    parameters:
      density: 1.0
      weight: 1.0
  - model: rombodawg/Rombos-LLM-V2.5-Qwen-72b # Continuous Fine-tuning 
    parameters:
      density: 1.0
      weight: 1.0    

merge_method: ties
base_model: Qwen/Qwen2.5-72B # Reflecting the most recent models as of 11/11/2024
parameters:
  normalize: true
  int8_mask: false
dtype: bfloat16
tokenizer_source: union # Experimental, preliminarily seems to garner good results
```