README.md

---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- MediaTek-Research/Breeze-7B-Instruct-v0.1
- Azure99/blossom-v4-mistral-7b
base_model:
- MediaTek-Research/Breeze-7B-Instruct-v0.1
- Azure99/blossom-v4-mistral-7b
---

# BreezePipe-7B-merge

BreezePipe-7B-merge is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [MediaTek-Research/Breeze-7B-Instruct-v0.1](https://huggingface.co/MediaTek-Research/Breeze-7B-Instruct-v0.1)
* [Azure99/blossom-v4-mistral-7b](https://huggingface.co/Azure99/blossom-v4-mistral-7b)

## 🧩 Configuration

```yaml
slices:
  - sources:
    - model: MediaTek-Research/Breeze-7B-Instruct-v0.1
      layer_range: [0, 32]
  - sources:
    - model: Azure99/blossom-v4-mistral-7b
      layer_range: [24, 32]
merge_method: passthrough
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Heng666/BreezePipe-7B-merge"
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"])
```