README.md

---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- Q-bert/MetaMath-Cybertron-Starling
- ozayezerceli/BetterSaul-7B-slerp
- chihoonlee10/T3Q-Mistral-Orca-Math-DPO
- EmbeddedLLM/Mistral-7B-Merge-14-v0.2
base_model:
- Q-bert/MetaMath-Cybertron-Starling
- ozayezerceli/BetterSaul-7B-slerp
- chihoonlee10/T3Q-Mistral-Orca-Math-DPO
- EmbeddedLLM/Mistral-7B-Merge-14-v0.2
---

# StrangeMerges_45-7B-dare_ties

StrangeMerges_45-7B-dare_ties is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Q-bert/MetaMath-Cybertron-Starling](https://huggingface.co/Q-bert/MetaMath-Cybertron-Starling)
* [ozayezerceli/BetterSaul-7B-slerp](https://huggingface.co/ozayezerceli/BetterSaul-7B-slerp)
* [chihoonlee10/T3Q-Mistral-Orca-Math-DPO](https://huggingface.co/chihoonlee10/T3Q-Mistral-Orca-Math-DPO)
* [EmbeddedLLM/Mistral-7B-Merge-14-v0.2](https://huggingface.co/EmbeddedLLM/Mistral-7B-Merge-14-v0.2)

## 🧩 Configuration

```yaml
models:
  - model: Q-bert/MetaMath-Cybertron-Starling
    parameters:
      weight: 0.3
      density: 0.53
  - model: ozayezerceli/BetterSaul-7B-slerp
    parameters:
      weight: 0.2
      density: 0.53
  - model: chihoonlee10/T3Q-Mistral-Orca-Math-DPO
    parameters:
      weight: 0.4
      density: 0.53
  - model: EmbeddedLLM/Mistral-7B-Merge-14-v0.2
    parameters:
      weight: 0.1
      density: 0.53
base_model: Gille/StrangeMerges_44-7B-dare_ties
merge_method: dare_ties
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Gille/StrangeMerges_45-7B-dare_ties"
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"])
```