README.md

---
tags:
- merge
- mergekit
- lazymergekit
- vicgalle/OpenHermes-Gemma-2B
- mlabonne/Gemmalpaca-2B
base_model:
- vicgalle/OpenHermes-Gemma-2B
- mlabonne/Gemmalpaca-2B
license: apache-2.0
---

# GemmaMerge-2B-Dare
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/64fc6d81d75293f417fee1d1/9GutFbLO3JMqAY2jQPJQQ.jpeg)
GemmaMerge-2B-Dare is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):

- [vicgalle/OpenHermes-Gemma-2B](https://huggingface.co/vicgalle/OpenHermes-Gemma-2B)
- [mlabonne/Gemmalpaca-2B](https://huggingface.co/mlabonne/Gemmalpaca-2B)

Special thanks to Charles Goddard for the quick implementation!

## 🏆 Evaluation
### Coming Soon

## 🧩 Configuration

```yaml
models:
  - model: vicgalle/OpenHermes-Gemma-2B
    parameters:
      density: 0.53
      weight: 0.5
  - model: mlabonne/Gemmalpaca-2B
    parameters:
      density: 0.53
      weight: 0.45
merge_method: dare_ties
base_model: vicgalle/OpenHermes-Gemma-2B
parameters:
  int8_mask: true
dtype: bfloat16

```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "johnsnowlabs/GemmaMerge-2B-Dare"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```