---
base_model:
- bunnycore/Phi-3.5-Hyper-Mix
- win10/phi3.5-mini-24-09-30
- bunnycore/Phi-3.5-Hyper-Mix
tags:
- merge
- mergekit
- lazymergekit
- bunnycore/Phi-3.5-Hyper-Mix
- win10/phi3.5-mini-24-09-30
---

# Phi-3.5-24-10-06

Phi-3.5-24-10-06 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [bunnycore/Phi-3.5-Hyper-Mix](https://huggingface.co/bunnycore/Phi-3.5-Hyper-Mix)
* [win10/phi3.5-mini-24-09-30](https://huggingface.co/win10/phi3.5-mini-24-09-30)
* [bunnycore/Phi-3.5-Hyper-Mix](https://huggingface.co/bunnycore/Phi-3.5-Hyper-Mix)

## 🧩 Configuration

```yaml
models:
  - model: bunnycore/Phi-3.5-Hyper-Mix
    parameters:
      density: 0.5
      weight: 0.5
  - model: win10/phi3.5-mini-24-09-30
    parameters:
      density: 0.5
      weight: 0.5
  - model: bunnycore/Phi-3.5-Hyper-Mix
    parameters:
      density: 0.5
      weight: 0.5
merge_method: ties
base_model: win10/Phi-3.5-mini-instruct-24-9-29
parameters:
  normalize: false
  int8_mask: true
dtype: float16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "win10/Phi-3.5-24-10-06"
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"])
```