File size: 1,717 Bytes
93a49ab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4bde0e6
 
93a49ab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
---
license: apache-2.0
tags:
- merge
- mergekit
- lazymergekit
- DreadPoor/Aurora_faustus-8B-LINEAR
- bunnycore/Llama-3.1-8B-TitanFusion-Mix
- Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base
- nvidia/OpenMath2-Llama3.1-8B
- vicgalle/Configurable-Llama-3.1-8B-Instruct
---

# Best-Mix-Llama-3.1-8B

This is a test merge with all my favorite models 

Best-Mix-Llama-3.1-8B is a merge of the following models using [mergekit](https://github.com/cg123/mergekit):
* [DreadPoor/Aurora_faustus-8B-LINEAR](https://huggingface.co./DreadPoor/Aurora_faustus-8B-LINEAR)
* [bunnycore/Llama-3.1-8B-TitanFusion-Mix](https://huggingface.co./bunnycore/Llama-3.1-8B-TitanFusion-Mix)
* [Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base](https://huggingface.co./Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base)
* [nvidia/OpenMath2-Llama3.1-8B](https://huggingface.co./nvidia/OpenMath2-Llama3.1-8B)
* [vicgalle/Configurable-Llama-3.1-8B-Instruct](https://huggingface.co./vicgalle/Configurable-Llama-3.1-8B-Instruct)

## 🧩 Configuration

```yaml
models:
  - model: DreadPoor/Aurora_faustus-8B-LINEAR
    parameters:
      density: 0.8
      weight: 0.8
  - model: bunnycore/Llama-3.1-8B-TitanFusion-Mix
    parameters:
      density: 0.8
      weight: 0.8
  - model: Joseph717171/Llama-3.1-SuperNova-8B-Lite_TIES_with_Base
    parameters:
      density: 0.8
      weight: 0.8
  - model: nvidia/OpenMath2-Llama3.1-8B
    parameters:
      density: 0.8
      weight: 0.8
  - model: vicgalle/Configurable-Llama-3.1-8B-Instruct
    parameters:
      density: 0.8
      weight: 0.8


merge_method: ties
base_model: bunnycore/Llama-3.1-8B-TitanFusion-Mix
parameters:
  normalize: false
  int8_mask: true
dtype: float16

```