README.md

---
tags:
- merge
- mergekit
- lazymergekit
- Kaoeiri/L3MaidRPKraiKei-V1.6-8B
- openlynn/Llama-3-Soliloquy-8B-v2
- Sao10K/L3-8B-Stheno-v3.2
base_model:
- Kaoeiri/L3MaidRPKraiKei-V1.6-8B
- openlynn/Llama-3-Soliloquy-8B-v2
- Sao10K/L3-8B-Stheno-v3.2
---

# L3MaidRPKraiKei-V1.65-8B

L3MaidRPKraiKei-V1.65-8B is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):

# Keep in mind that, this merged model isn't usually tested at the moment, which could benefit in vocabulary error.
* [Kaoeiri/L3MaidRPKraiKei-V1.6-8B](https://huggingface.co/Kaoeiri/L3MaidRPKraiKei-V1.6-8B)
* [openlynn/Llama-3-Soliloquy-8B-v2](https://huggingface.co/openlynn/Llama-3-Soliloquy-8B-v2)
* [Sao10K/L3-8B-Stheno-v3.2](https://huggingface.co/Sao10K/L3-8B-Stheno-v3.2)

## 🧩 Configuration

```yaml
models:
  - model: Kaoeiri/L3MaidRPKraiKei-V1.6-8B
    parameters:
      density: .5
      weight: 1
  - model: openlynn/Llama-3-Soliloquy-8B-v2
    parameters:
      density: .35
      weight: .4
  - model: Sao10K/L3-8B-Stheno-v3.2
    parameters:
      density: .5
      weight: .7
merge_method: ties

base_model: mlabonne/NeuralDaredevil-8B-abliterated
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Kaoeiri/L3MaidRPKraiKei-V1.65-8B"
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"])
```