File size: 2,614 Bytes
413e31d
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
---
base_model:
- migtissera/Tess-3-Llama-3.1-70B
- HODACHI/Llama-3.1-70B-EZO-1.1-it
- shenzhi-wang/Llama3.1-70B-Chinese-Chat
- Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-dpo-70B
tags:
- merge
- mergekit
- lazymergekit
- migtissera/Tess-3-Llama-3.1-70B
- HODACHI/Llama-3.1-70B-EZO-1.1-it
- shenzhi-wang/Llama3.1-70B-Chinese-Chat
- Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-dpo-70B
---

# L3.1-70b-MeowMix

Meow.

L3.1-70b-MeowMix is a merge of the following models using [LazyMergekit](https://colab.research.google.com/drive/1obulZ1ROXHjYLn6PPZJwRR6GzgQogxxb?usp=sharing) running on Runpod:
* [migtissera/Tess-3-Llama-3.1-70B](https://huggingface.co/migtissera/Tess-3-Llama-3.1-70B)
* [HODACHI/Llama-3.1-70B-EZO-1.1-it](https://huggingface.co/HODACHI/Llama-3.1-70B-EZO-1.1-it)
* [shenzhi-wang/Llama3.1-70B-Chinese-Chat](https://huggingface.co/shenzhi-wang/Llama3.1-70B-Chinese-Chat)
* [Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-dpo-70B](https://huggingface.co/Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-dpo-70B)

## Yap / Chat Format

Llama 3 Instruct.

## 🧩 Configuration

```yaml

# Thanks to nruaif for suggesting MeowMix merge part.
# Merge CJK into migtissera/Tess-3-Llama-3.1-70B

models:
  - model: migtissera/Tess-3-Llama-3.1-70B
    parameters:
      density: 0.7
      weight:
        - value: 0.75
  - model: HODACHI/Llama-3.1-70B-EZO-1.1-it
    parameters:
      density: 0.2
      weight:
        - value: [1, 0.75, 0.5, 0.25, 0, 0, 0, 0, 0.0, 0.5, 1]
  - model: shenzhi-wang/Llama3.1-70B-Chinese-Chat
    parameters:
      density: 0.2
      weight:
        - value: [1, 0.75, 0.5, 0.25, 0, 0, 0, 0, 0.0, 0.5, 1]
  - model: Saxo/Linkbricks-Horizon-AI-Korean-llama3.1-sft-dpo-70B
    parameters:
      density: 0.2
      weight:
        - value: [1, 0.75, 0.5, 0.25, 0, 0, 0, 0, 0.0, 0.5, 1]

merge_method: ties
base_model: migtissera/Tess-3-Llama-3.1-70B
parameters:
  normalize: true
dtype: bfloat16
```

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "KaraKaraWitch/L3.1-70b-MeowMix"
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"])
```