File size: 2,945 Bytes
670e3bf ff99dd1 670e3bf 5542217 670e3bf |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 |
---
tags:
- merge
- mergekit
- lazymergekit
- SanjiWatsuki/Kunoichi-DPO-v2-7B
- eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
base_model:
- SanjiWatsuki/Kunoichi-DPO-v2-7B
- eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
license: cc-by-nc-2.0
---
# kuno-royale-v2-7b
An attempt to further strengthen the roleplaying prose of [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co./SanjiWatsuki/Kunoichi-DPO-v2-7B) using [eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO](https://huggingface.co./eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO), a high-scorer for 7B models on the [Open LLM Leaderboard](https://huggingface.co./spaces/HuggingFaceH4/open_llm_leaderboard).
Personal RP tests prove promising, and meaningless leaderboard metrics have improved vs [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co./SanjiWatsuki/Kunoichi-DPO-v2-7B).
|Model | Average | ARC | HellaSwag | MMLU | TruthfulQA | Winogrande | GSM8K |
|-------------------|---------|-----|-----------|------|------------|------------|-------|
| eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO | 76.45 | 73.12 | 89.09 | 64.80 | 77.45 | 84.77 | 69.45 |
| **core-3/kuno-royale-v2-7b** | **74.80** | **72.01** | **88.15** | **65.07** | **71.10** | **82.24** | **70.20** |
| SanjiWatsuki/Kunoichi-DPO-v2-7B | 72.46 | 69.62 | 87.44 | 64.94 | 66.06 | 80.82 | 65.88 |
| SanjiWatsuki/Kunoichi-7B | 72.13 | 68.69 | 87.10 | 64.90 | 64.04 | 81.06 | 67.02 |
kuno-royale-v2-7b is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing):
* [SanjiWatsuki/Kunoichi-DPO-v2-7B](https://huggingface.co./SanjiWatsuki/Kunoichi-DPO-v2-7B)
* [eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO](https://huggingface.co./eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO)
## 🧩 Configuration
```yaml
slices:
- sources:
- model: SanjiWatsuki/Kunoichi-DPO-v2-7B
layer_range: [0, 32]
- model: eren23/ogno-monarch-jaskier-merge-7b-OH-PREF-DPO
layer_range: [0, 32]
merge_method: slerp
base_model: SanjiWatsuki/Kunoichi-DPO-v2-7B
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: bfloat16
```
## 💻 Usage
```python
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "core-3/kuno-royale-v2-7b"
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"])
``` |