Update README.md
Browse files
README.md
CHANGED
@@ -1,5 +1,103 @@
|
|
1 |
-
---
|
2 |
-
|
3 |
-
|
4 |
-
|
5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
base_model:
|
3 |
+
- NousResearch/Hermes-2-Pro-Llama-3-8B
|
4 |
+
- cognitivecomputations/dolphin-2.9-llama3-8b
|
5 |
+
- NousResearch/Meta-Llama-3-8B
|
6 |
+
- winglian/llama-3-8b-256k-PoSE
|
7 |
+
- maum-ai/Llama-3-MAAL-8B-Instruct-v0.1
|
8 |
+
- asiansoul/Llama-3-Open-Ko-Linear-8B
|
9 |
+
- NousResearch/Meta-Llama-3-8B-Instruct
|
10 |
+
- nvidia/Llama3-ChatQA-1.5-8B
|
11 |
+
- Danielbrdz/Barcenas-Llama3-8b-ORPO
|
12 |
+
- aaditya/Llama3-OpenBioLLM-8B
|
13 |
+
library_name: transformers
|
14 |
+
tags:
|
15 |
+
- mergekit
|
16 |
+
- merge
|
17 |
+
- llama
|
18 |
+
|
19 |
+
---
|
20 |
+
# YACHT-Llama-3-Ko-8B
|
21 |
+
|
22 |
+
[![DALL-E Yacht](https://i.ibb.co/hHr5xnh/DALL-E-2024-05-05-11-57-02-A-futuristic-yacht-boat-on-a-calm-ocean-at-dawn-featuring-sleek-curves-an.png)](https://ibb.co/92BXmfz)
|
23 |
+
|
24 |
+
## Merge Details
|
25 |
+
### Merge Method
|
26 |
+
|
27 |
+
This model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [NousResearch/Meta-Llama-3-8B](https://huggingface.co/NousResearch/Meta-Llama-3-8B) as a base.
|
28 |
+
|
29 |
+
### Models Merged
|
30 |
+
|
31 |
+
The following models were included in the merge:
|
32 |
+
* [NousResearch/Hermes-2-Pro-Llama-3-8B](https://huggingface.co/NousResearch/Hermes-2-Pro-Llama-3-8B)
|
33 |
+
* [cognitivecomputations/dolphin-2.9-llama3-8b](https://huggingface.co/cognitivecomputations/dolphin-2.9-llama3-8b)
|
34 |
+
* [winglian/llama-3-8b-256k-PoSE](https://huggingface.co/winglian/llama-3-8b-256k-PoSE)
|
35 |
+
* [maum-ai/Llama-3-MAAL-8B-Instruct-v0.1](https://huggingface.co/maum-ai/Llama-3-MAAL-8B-Instruct-v0.1)
|
36 |
+
* [asiansoul/Llama-3-Open-Ko-Linear-8B](https://huggingface.co/asiansoul/Llama-3-Open-Ko-Linear-8B)
|
37 |
+
* [NousResearch/Meta-Llama-3-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct)
|
38 |
+
* [nvidia/Llama3-ChatQA-1.5-8B](https://huggingface.co/nvidia/Llama3-ChatQA-1.5-8B)
|
39 |
+
* [Danielbrdz/Barcenas-Llama3-8b-ORPO](https://huggingface.co/Danielbrdz/Barcenas-Llama3-8b-ORPO)
|
40 |
+
* [aaditya/Llama3-OpenBioLLM-8B](https://huggingface.co/aaditya/Llama3-OpenBioLLM-8B)
|
41 |
+
|
42 |
+
### Configuration
|
43 |
+
|
44 |
+
The following YAML configuration was used to produce this model:
|
45 |
+
|
46 |
+
```yaml
|
47 |
+
models:
|
48 |
+
- model: NousResearch/Meta-Llama-3-8B
|
49 |
+
# Base model providing a general foundation without specific parameters
|
50 |
+
|
51 |
+
- model: NousResearch/Meta-Llama-3-8B-Instruct
|
52 |
+
parameters:
|
53 |
+
density: 0.60
|
54 |
+
weight: 0.25
|
55 |
+
|
56 |
+
- model: winglian/llama-3-8b-256k-PoSE
|
57 |
+
parameters:
|
58 |
+
density: 0.55
|
59 |
+
weight: 0.15
|
60 |
+
|
61 |
+
- model: nvidia/Llama3-ChatQA-1.5-8B
|
62 |
+
parameters:
|
63 |
+
density: 0.55
|
64 |
+
weight: 0.1
|
65 |
+
|
66 |
+
- model: asiansoul/Llama-3-Open-Ko-Linear-8B
|
67 |
+
parameters:
|
68 |
+
density: 0.55
|
69 |
+
weight: 0.2
|
70 |
+
|
71 |
+
- model: maum-ai/Llama-3-MAAL-8B-Instruct-v0.1
|
72 |
+
parameters:
|
73 |
+
density: 0.55
|
74 |
+
weight: 0.1
|
75 |
+
|
76 |
+
- model: NousResearch/Hermes-2-Pro-Llama-3-8B
|
77 |
+
parameters:
|
78 |
+
density: 0.55
|
79 |
+
weight: 0.1
|
80 |
+
|
81 |
+
- model: cognitivecomputations/dolphin-2.9-llama3-8b
|
82 |
+
parameters:
|
83 |
+
density: 0.55
|
84 |
+
weight: 0.05
|
85 |
+
|
86 |
+
- model: Danielbrdz/Barcenas-Llama3-8b-ORPO
|
87 |
+
parameters:
|
88 |
+
density: 0.55
|
89 |
+
weight: 0.05
|
90 |
+
|
91 |
+
- model: aaditya/Llama3-OpenBioLLM-8B
|
92 |
+
parameters:
|
93 |
+
density: 0.55
|
94 |
+
weight: 0.1
|
95 |
+
|
96 |
+
merge_method: dare_ties
|
97 |
+
base_model: NousResearch/Meta-Llama-3-8B
|
98 |
+
parameters:
|
99 |
+
int8_mask: true
|
100 |
+
dtype: bfloat16
|
101 |
+
|
102 |
+
|
103 |
+
```
|