munish0838
commited on
Commit
•
8975a1f
1
Parent(s):
fc2aeec
Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,147 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
tags:
|
4 |
+
- llama-factory
|
5 |
+
license: llama3
|
6 |
+
datasets:
|
7 |
+
- TheSkullery/Aether-Lite-V1.2
|
8 |
+
base_model: Steelskull/L3-Aethora-15B
|
9 |
+
pipeline_tag: text-generation
|
10 |
+
---
|
11 |
+
|
12 |
+
# QuantFactory/L3-Aethora-15B-GGUF
|
13 |
+
This is quantized version of [Steelskull/L3-Aethora-15B](https://huggingface.co/Steelskull/L3-Aethora-15B) created using llama.cpp
|
14 |
+
|
15 |
+
# Model Description
|
16 |
+
|
17 |
+
<!DOCTYPE html>
|
18 |
+
<style>
|
19 |
+
body, html {
|
20 |
+
height: 100%; /* Ensure the full height of the page is used */
|
21 |
+
margin: 0;
|
22 |
+
padding: 0;
|
23 |
+
font-family: 'Quicksand', sans-serif;
|
24 |
+
background: linear-gradient(135deg, #2E3440 0%, #1A202C 100%);
|
25 |
+
color: #D8DEE9;
|
26 |
+
font-size: 16px;
|
27 |
+
}
|
28 |
+
|
29 |
+
.container {
|
30 |
+
width: 100%; /* Full width */
|
31 |
+
height: 100%; /* Full height */
|
32 |
+
padding: 20px;
|
33 |
+
margin: 0; /* Remove margin to fill the entire area */
|
34 |
+
background-color: rgba(255, 255, 255, 0.02);
|
35 |
+
border-radius: 12px;
|
36 |
+
box-shadow: 0 4px 10px rgba(0, 0, 0, 0.2);
|
37 |
+
backdrop-filter: blur(10px);
|
38 |
+
border: 1px solid rgba(255, 255, 255, 0.1);
|
39 |
+
}
|
40 |
+
|
41 |
+
.header h1 {
|
42 |
+
font-size: 28px;
|
43 |
+
color: #5F9EA0;
|
44 |
+
margin: 0 0 20px 0;
|
45 |
+
text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);
|
46 |
+
}
|
47 |
+
|
48 |
+
.update-section h2 {
|
49 |
+
font-size: 24px;
|
50 |
+
color: #88C0D0;
|
51 |
+
}
|
52 |
+
|
53 |
+
.update-section p {
|
54 |
+
font-size: 16px;
|
55 |
+
line-height: 1.6;
|
56 |
+
color: #ECEFF4;
|
57 |
+
}
|
58 |
+
|
59 |
+
.info img {
|
60 |
+
width: 100%;
|
61 |
+
border-radius: 10px;
|
62 |
+
margin-bottom: 15px;
|
63 |
+
}
|
64 |
+
|
65 |
+
a {
|
66 |
+
color: #88C0D0;
|
67 |
+
text-decoration: none;
|
68 |
+
}
|
69 |
+
|
70 |
+
a:hover {
|
71 |
+
color: #A3BE8C;
|
72 |
+
}
|
73 |
+
|
74 |
+
.button {
|
75 |
+
display: inline-block;
|
76 |
+
background-color: #5E81AC;
|
77 |
+
color: #E5E9F0;
|
78 |
+
padding: 10px 20px;
|
79 |
+
border-radius: 5px;
|
80 |
+
cursor: pointer;
|
81 |
+
text-decoration: none;
|
82 |
+
}
|
83 |
+
|
84 |
+
.button:hover {
|
85 |
+
background-color: #81A1C1;
|
86 |
+
}
|
87 |
+
|
88 |
+
pre {
|
89 |
+
background-color: #2E3440;
|
90 |
+
padding: 10px;
|
91 |
+
border-radius: 5px;
|
92 |
+
overflow-x: auto;
|
93 |
+
}
|
94 |
+
|
95 |
+
code {
|
96 |
+
font-family: 'Courier New', monospace;
|
97 |
+
color: #D8DEE9;
|
98 |
+
}
|
99 |
+
</style>
|
100 |
+
|
101 |
+
<html lang="en">
|
102 |
+
<head>
|
103 |
+
<meta charset="UTF-8">
|
104 |
+
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
105 |
+
<title>L3-Aethora-15B Data Card</title>
|
106 |
+
<link href="https://fonts.googleapis.com/css2?family=Quicksand:wght@400;500;600&display=swap" rel="stylesheet">
|
107 |
+
</head>
|
108 |
+
<body>
|
109 |
+
<div class="container">
|
110 |
+
<div class="header">
|
111 |
+
<h1>L3-Aethora-15B</h1>
|
112 |
+
</div>
|
113 |
+
<div class="info">
|
114 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/64545af5ec40bbbd01242ca6/W0qzZK_V1Zt1GdgCIsnrP.png">
|
115 |
+
<p>The Skullery Presents L3-Aethora-15B.</p>
|
116 |
+
<p><strong>Creator:</strong> <a href="https://huggingface.co/steelskull" target="_blank">Steelskull</a></p>
|
117 |
+
<p><strong>Dataset:</strong> <a href="https://huggingface.co/datasets/TheSkullery/Aether-Lite-V1.2" target="_blank">Aether-Lite-V1.2</a></p>
|
118 |
+
<p><strong>Trained:</strong> 4 x A100 for 15 hours Using RsLora and DORA</p>
|
119 |
+
<h1>About L3-Aethora-15B:</h1>
|
120 |
+
<pre><code> L3 = Llama3 </code></pre>
|
121 |
+
<p>L3-Aethora-15B was crafted through using the abilteration method to adjust model responses. The model's refusal is inhibited, focusing on yielding more compliant and facilitative dialogue interactions. It then underwent a modified DUS (Depth Up Scale) merge (originally used by @Elinas) by using passthrough merge to create a 15b model, with specific adjustments (zeroing) to 'o_proj' and 'down_proj', enhancing its efficiency and reducing perplexity. This created AbL3In-15b.<br>
|
122 |
+
<p>AbL3In-15b was then trained for 4 epochs using Rslora & DORA training methods on the Aether-Lite-V1.2 dataset, containing ~82000 high quality samples, designed to strike a fine balance between creativity, slop, and intelligence at about a 60/40 split</p>
|
123 |
+
<p>This model is trained on the L3 prompt format.</p>
|
124 |
+
<h2>Quants:</h2>
|
125 |
+
<li><a href="https://huggingface.co/mradermacher/L3-Aethora-15B-GGUF" target="_blank">Mradermacher/L3-Aethora-15B-GGUF</a></li>
|
126 |
+
<li><a href="https://huggingface.co/mradermacher/L3-Aethora-15B-i1-GGUF" target="_blank">Mradermacher/L3-Aethora-15B-i1-GGUF</a></li>
|
127 |
+
<li><a href="https://huggingface.co/NikolayKozloff" target="_blank">NikolayKozloff/L3-Aethora-15B-GGUF</a></li>
|
128 |
+
<p></p>
|
129 |
+
<h2>Dataset Summary: (Filtered)</h2>
|
130 |
+
<p>Filtered Phrases: GPTslop, Claudism's</p>
|
131 |
+
<ul>
|
132 |
+
<li><strong>mrfakename/Pure-Dove-ShareGPT:</strong> Processed 3707, Removed 150</li>
|
133 |
+
<li><strong>mrfakename/Capybara-ShareGPT:</strong> Processed 13412, Removed 2594</li>
|
134 |
+
<li><strong>jondurbin/airoboros-3.2:</strong> Processed 54517, Removed 4192</li>
|
135 |
+
<li><strong>PJMixers/grimulkan_theory-of-mind-ShareGPT:</strong> Processed 533, Removed 6</li>
|
136 |
+
<li><strong>grimulkan/PIPPA-augmented-dedup:</strong> Processed 869, Removed 46</li>
|
137 |
+
<li><strong>grimulkan/LimaRP-augmented:</strong> Processed 790, Removed 14</li>
|
138 |
+
<li><strong>PJMixers/grimulkan_physical-reasoning-ShareGPT:</strong> Processed 895, Removed 4</li>
|
139 |
+
<li><strong>MinervaAI/Aesir-Preview:</strong> Processed 994, Removed 6</li>
|
140 |
+
<li><strong>Doctor-Shotgun/no-robots-sharegpt:</strong> Processed 9911, Removed 89</li>
|
141 |
+
</ul>
|
142 |
+
<h2>Deduplication Stats:</h2>
|
143 |
+
<p>Starting row count: 85628, Final row count: 81960, Rows removed: 3668</p>
|
144 |
+
</div>
|
145 |
+
</div>
|
146 |
+
</body>
|
147 |
+
</html>
|