README.md

---
tags:
- merge
- mergekit
- lazymergekit
- abacaj/phi-2-super
- mobiuslabsgmbh/aanaphi2-v0.1
base_model:
- abacaj/phi-2-super
- mobiuslabsgmbh/aanaphi2-v0.1
license: apache-2.0
---

# PhiMerge-2.7B-Dare
![image/png](https://cdn-uploads.huggingface.co/production/uploads/660cfe98280a82e38fe4ef49/rcSJbgdC-9F9MyUwKkhEb.png)
PhiMerge-2.7B-Dare is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):

* [abacaj/phi-2-super](https://huggingface.co/abacaj/phi-2-super)
* [mobiuslabsgmbh/aanaphi2-v0.1](https://huggingface.co/mobiuslabsgmbh/aanaphi2-v0.1)

## 🏆 Evaluation results
### Coming Soon

## 🧩 Configuration

```yaml
models:
  - model: microsoft/phi-2
    # No parameters necessary for base model
  - model: abacaj/phi-2-super
    parameters:
      density: 0.53
      weight: 0.5
  - model: mobiuslabsgmbh/aanaphi2-v0.1
    parameters:
      density: 0.53
      weight: 0.45
merge_method: dare_ties
base_model: microsoft/phi-2
parameters:
  int8_mask: true
dtype: float16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "johnsnowlabs/PhiMerge-2.7B-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"])
```