README.md

---
tags:
- merge
- mergekit
- lazymergekit
- aipib/Tinyllama-moe3
- aipib/karasu-moe
base_model:
- aipib/Tinyllama-moe3
- aipib/karasu-moe
---

# karasu-moexdareties

karasu-moexdareties is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [aipib/Tinyllama-moe3](https://huggingface.co/aipib/Tinyllama-moe3)
* [aipib/karasu-moe](https://huggingface.co/aipib/karasu-moe)

## 🧩 Configuration

```yaml
slices:
- sources:
  - layer_range: [0, 22]
    model: aipib/Tinyllama-moe3
    parameters:
      density: [1, 0.7, 0.1]
      weight: 1.0
  - layer_range: [0, 22]
    model: aipib/karasu-moe
    parameters:
      density: 0.33
      weight:
        - filter: mlp
          value: 0.5
        - value: 0
merge_method: dare_ties
base_model: aipib/Tinyllama-moe3
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "aipib/karasu-moexdareties"
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"])
```