README.md

---
base_model:
- Weyaxi/MetaMath-OpenHermes-2.5-neural-chat-7b-v3-1-7B-Linear
- icefog72/IceMoonshineRP-7b
- Weyaxi/MetaMath-neural-chat-7b-v3-2-Slerp
- VAGOsolutions/SauerkrautLM-7b-HerO
- mrfakename/NeuralOrca-7B-v1
tags:
- merge
- mergekit
- lazymergekit
- Weyaxi/MetaMath-OpenHermes-2.5-neural-chat-7b-v3-1-7B-Linear
- icefog72/IceMoonshineRP-7b
- Weyaxi/MetaMath-neural-chat-7b-v3-2-Slerp
- VAGOsolutions/SauerkrautLM-7b-HerO
- mrfakename/NeuralOrca-7B-v1
---

# kangaroo_7B_test01

kangaroo_7B_test01 is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [Weyaxi/MetaMath-OpenHermes-2.5-neural-chat-7b-v3-1-7B-Linear](https://huggingface.co/Weyaxi/MetaMath-OpenHermes-2.5-neural-chat-7b-v3-1-7B-Linear)
* [icefog72/IceMoonshineRP-7b](https://huggingface.co/icefog72/IceMoonshineRP-7b)
* [Weyaxi/MetaMath-neural-chat-7b-v3-2-Slerp](https://huggingface.co/Weyaxi/MetaMath-neural-chat-7b-v3-2-Slerp)
* [VAGOsolutions/SauerkrautLM-7b-HerO](https://huggingface.co/VAGOsolutions/SauerkrautLM-7b-HerO)
* [mrfakename/NeuralOrca-7B-v1](https://huggingface.co/mrfakename/NeuralOrca-7B-v1)

## 🧩 Configuration

```yaml
models:
  - model: BioMistral/BioMistral-7B-DARE
    # No parameters necessary for base model
  - model: Weyaxi/MetaMath-OpenHermes-2.5-neural-chat-7b-v3-1-7B-Linear
    parameters:
      density: 0.5
      weight: 0.2


  - model: icefog72/IceMoonshineRP-7b
    parameters:
      density: 0.5
      weight: 0.2


  - model: Weyaxi/MetaMath-neural-chat-7b-v3-2-Slerp
    parameters:
      density: 0.5
      weight: 0.2


  - model: VAGOsolutions/SauerkrautLM-7b-HerO
    parameters:
      density: 0.5
      weight: 0.2


  - model: mrfakename/NeuralOrca-7B-v1
    parameters:
      density: 0.5
      weight: 0.2


merge_method: dare_ties
base_model: BioMistral/BioMistral-7B-DARE
parameters:
  int8_mask: true
dtype: bfloat16

```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "kainatq/kangaroo_7B_test01"
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"])
```