Steelskull
commited on
Commit
•
3b3083b
1
Parent(s):
e4f0883
Update README.md
Browse files
README.md
CHANGED
@@ -5,38 +5,145 @@ base_model:
|
|
5 |
- TheDrummer/Gemmasutra-Pro-27B-v1
|
6 |
library_name: transformers
|
7 |
tags:
|
8 |
-
- mergekit
|
9 |
- merge
|
10 |
|
11 |
---
|
12 |
-
|
13 |
-
|
14 |
-
|
15 |
-
|
16 |
-
## Merge Details
|
17 |
-
### Merge Method
|
18 |
-
|
19 |
-
This model was merged using the della merge method using [TheDrummer/Big-Tiger-Gemma-27B-v1](https://huggingface.co/TheDrummer/Big-Tiger-Gemma-27B-v1) as a base.
|
20 |
-
|
21 |
-
### Models Merged
|
22 |
-
|
23 |
-
The following models were included in the merge:
|
24 |
-
* [migtissera/Tess-v2.5-Gemma-2-27B-alpha](https://huggingface.co/migtissera/Tess-v2.5-Gemma-2-27B-alpha)
|
25 |
-
* [TheDrummer/Gemmasutra-Pro-27B-v1](https://huggingface.co/TheDrummer/Gemmasutra-Pro-27B-v1)
|
26 |
-
|
27 |
-
### Configuration
|
28 |
-
|
29 |
-
The following YAML configuration was used to produce this model:
|
30 |
|
31 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
base_model: TheDrummer/Big-Tiger-Gemma-27B-v1
|
33 |
merge_method: della
|
34 |
dtype: bfloat16
|
35 |
models:
|
36 |
- model: TheDrummer/Gemmasutra-Pro-27B-v1
|
37 |
-
parameters:
|
38 |
-
weight: 1.0
|
39 |
- model: migtissera/Tess-v2.5-Gemma-2-27B-alpha
|
40 |
-
|
41 |
-
|
42 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5 |
- TheDrummer/Gemmasutra-Pro-27B-v1
|
6 |
library_name: transformers
|
7 |
tags:
|
|
|
8 |
- merge
|
9 |
|
10 |
---
|
11 |
+
---
|
12 |
+
license: apache-2.0
|
13 |
+
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
14 |
|
15 |
+
<!DOCTYPE html>
|
16 |
+
<html lang="en">
|
17 |
+
<head>
|
18 |
+
<meta charset="UTF-8">
|
19 |
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
20 |
+
<title>G2-MS-Nyxora-27b Data Card</title>
|
21 |
+
<link href="https://fonts.googleapis.com/css2?family=Quicksand:wght@400;500;600&display=swap" rel="stylesheet">
|
22 |
+
<style>
|
23 |
+
body, html {
|
24 |
+
height: 100%;
|
25 |
+
margin: 0;
|
26 |
+
padding: 0;
|
27 |
+
font-family: 'Quicksand', sans-serif;
|
28 |
+
background: linear-gradient(135deg, #0a1128 0%, #1c2541 100%);
|
29 |
+
color: #e0e1dd;
|
30 |
+
font-size: 16px;
|
31 |
+
}
|
32 |
+
.container {
|
33 |
+
width: 100%;
|
34 |
+
height: 100%;
|
35 |
+
padding: 20px;
|
36 |
+
margin: 0;
|
37 |
+
background-color: rgba(255, 255, 255, 0.05);
|
38 |
+
border-radius: 12px;
|
39 |
+
box-shadow: 0 4px 10px rgba(0, 0, 0, 0.3);
|
40 |
+
backdrop-filter: blur(10px);
|
41 |
+
border: 1px solid rgba(255, 255, 255, 0.1);
|
42 |
+
}
|
43 |
+
.header h1 {
|
44 |
+
font-size: 28px;
|
45 |
+
color: #4cc9f0;
|
46 |
+
margin: 0 0 20px 0;
|
47 |
+
text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);
|
48 |
+
}
|
49 |
+
.update-section h2 {
|
50 |
+
font-size: 24px;
|
51 |
+
color: #7209b7;
|
52 |
+
}
|
53 |
+
.update-section p {
|
54 |
+
font-size: 16px;
|
55 |
+
line-height: 1.6;
|
56 |
+
color: #e0e1dd;
|
57 |
+
}
|
58 |
+
.info img {
|
59 |
+
width: 100%;
|
60 |
+
border-radius: 10px;
|
61 |
+
margin-bottom: 15px;
|
62 |
+
}
|
63 |
+
a {
|
64 |
+
color: #4cc9f0;
|
65 |
+
text-decoration: none;
|
66 |
+
}
|
67 |
+
a:hover {
|
68 |
+
color: #f72585;
|
69 |
+
}
|
70 |
+
.button {
|
71 |
+
display: inline-block;
|
72 |
+
background-color: #3a0ca3;
|
73 |
+
color: #e0e1dd;
|
74 |
+
padding: 10px 20px;
|
75 |
+
border-radius: 5px;
|
76 |
+
cursor: pointer;
|
77 |
+
text-decoration: none;
|
78 |
+
}
|
79 |
+
.button:hover {
|
80 |
+
background-color: #7209b7;
|
81 |
+
}
|
82 |
+
pre {
|
83 |
+
background-color: #1c2541;
|
84 |
+
padding: 10px;
|
85 |
+
border-radius: 5px;
|
86 |
+
overflow-x: auto;
|
87 |
+
}
|
88 |
+
code {
|
89 |
+
font-family: 'Courier New', monospace;
|
90 |
+
color: #e0e1dd;
|
91 |
+
}
|
92 |
+
.donation-section {
|
93 |
+
margin-top: 20px;
|
94 |
+
text-align: center;
|
95 |
+
}
|
96 |
+
.donation-section p {
|
97 |
+
margin-bottom: 10px;
|
98 |
+
}
|
99 |
+
</style>
|
100 |
+
</head>
|
101 |
+
<body>
|
102 |
+
<div class="container">
|
103 |
+
<div class="header">
|
104 |
+
<h1>G2-DA-Nyxora-27b-V2</h1>
|
105 |
+
</div>
|
106 |
+
<div class="info">
|
107 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/64545af5ec40bbbd01242ca6/rqcAEjzvLpQwu2Cg3g1hH.png" alt="Model Image">
|
108 |
+
<p>Now that the cute anime girl has your attention.</p>
|
109 |
+
<p><strong>Creator:</strong> <a href="https://huggingface.co/Steelskull" target="_blank">SteelSkull</a></p>
|
110 |
+
<h1>About G2-DA-Nyxora-27b-V2:</h1>
|
111 |
+
<pre><code><strong>Model Name Legend =</strong>
|
112 |
+
"G2 = Gemma 2"
|
113 |
+
"DA = Della"
|
114 |
+
"Nyxora = Tiger Base = cat = Nya"
|
115 |
+
"27b = its 27b"
|
116 |
+
"V2 = V2"</code></pre>
|
117 |
+
<p>This model is the second experimental foray into 27b models. Feedback is welcome for further improvements.</p>
|
118 |
+
<p>G2-DA-Nyxora-27b-V2 combines multiple models' strengths to provide a versatile assistant for various tasks, including general use, storytelling, and roleplay (ERP & RP). </p>
|
119 |
+
<p>The Model seems to have fixed alot of the prudish issues of the last model.</p>
|
120 |
+
<p>The new Exprimental merge method Della builds upon DARE, It uses adaptive pruning based on parameter magnitudes. DELLA first ranks parameters in each row of delta parameters and assigns drop probabilities inversely proportional to their magnitudes. This allows it to retain more important changes while reducing interference. After pruning, it rescales the remaining parameters similar to DARE.</p>
|
121 |
+
<h2>Quants:</h2>
|
122 |
+
<p>Will add as found or im notified of their creation (If you know of one create a discussion!)
|
123 |
+
<li><a href="https://huggingface.co/SteelQuants">SteelQuants</a>, My Personal Quants</li>
|
124 |
+
<p></p>
|
125 |
+
<h2>Config:</h2>
|
126 |
+
<pre><code>MODEL_NAME = "G2-DA-Nyxora-27b-V2"
|
127 |
+
yaml_config = """
|
128 |
base_model: TheDrummer/Big-Tiger-Gemma-27B-v1
|
129 |
merge_method: della
|
130 |
dtype: bfloat16
|
131 |
models:
|
132 |
- model: TheDrummer/Gemmasutra-Pro-27B-v1
|
|
|
|
|
133 |
- model: migtissera/Tess-v2.5-Gemma-2-27B-alpha
|
134 |
+
"""</code></pre>
|
135 |
+
<h4>Template:</h4>
|
136 |
+
<pre><code><start_of_turn>user
|
137 |
+
{{ if .System }}{{ .System }} {{ end }}{{ .Prompt }}<end_of_turn>
|
138 |
+
<start_of_turn>model
|
139 |
+
{{ .Response }}<end_of_turn></code></pre>
|
140 |
+
<p><strong>I've had a few people ask about donations so here's a link:</strong</p>
|
141 |
+
</div>
|
142 |
+
<div class="donation-section">
|
143 |
+
<a href="https://ko-fi.com/Y8Y0AO2XE" target="_blank">
|
144 |
+
<img height="36" style="border:0px;height:36px;" src="https://storage.ko-fi.com/cdn/kofi2.png?v=3" border="0" alt="Buy Me a Coffee at ko-fi.com" />
|
145 |
+
</a>
|
146 |
+
</div>
|
147 |
+
</div>
|
148 |
+
</body>
|
149 |
+
</html>
|