mikeumus-divincian commited on
Commit
8b5a46d
·
verified ·
1 Parent(s): c8f0aac

Add org card HTML content

Browse files
Files changed (1) hide show
  1. index.html +68 -19
index.html CHANGED
@@ -1,19 +1,68 @@
1
- <!doctype html>
2
- <html>
3
- <head>
4
- <meta charset="utf-8" />
5
- <meta name="viewport" content="width=device-width" />
6
- <title>My static Space</title>
7
- <link rel="stylesheet" href="style.css" />
8
- </head>
9
- <body>
10
- <div class="card">
11
- <h1>Welcome to your static Space!</h1>
12
- <p>You can modify this app directly by editing <i>index.html</i> in the Files and versions tab.</p>
13
- <p>
14
- Also don't forget to check the
15
- <a href="https://huggingface.co/docs/hub/spaces" target="_blank">Spaces documentation</a>.
16
- </p>
17
- </div>
18
- </body>
19
- </html>
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <!DOCTYPE html>
2
+ <html lang="en">
3
+ <head>
4
+ <meta charset="UTF-8">
5
+ <meta name="viewport" content="width=device-width, initial-scale=1.0">
6
+ <title>Divinci AI</title>
7
+ <style>
8
+ body { font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', sans-serif; max-width: 860px; margin: 0 auto; padding: 2rem 1.5rem; color: #1a1a1a; line-height: 1.6; }
9
+ h1 { font-size: 1.8rem; font-weight: 700; margin-bottom: 0.25rem; }
10
+ h2 { font-size: 1.1rem; font-weight: 600; margin-top: 2rem; margin-bottom: 0.5rem; border-bottom: 1px solid #e5e7eb; padding-bottom: 0.25rem; }
11
+ p { margin: 0.5rem 0 1rem; }
12
+ table { border-collapse: collapse; width: 100%; margin: 1rem 0; font-size: 0.9rem; }
13
+ th, td { border: 1px solid #e5e7eb; padding: 0.5rem 0.75rem; text-align: left; }
14
+ th { background: #f9fafb; font-weight: 600; }
15
+ a { color: #2563eb; text-decoration: none; }
16
+ a:hover { text-decoration: underline; }
17
+ .tagline { color: #6b7280; font-size: 1rem; margin-bottom: 1.5rem; }
18
+ .footer { margin-top: 2.5rem; padding-top: 1rem; border-top: 1px solid #e5e7eb; font-size: 0.85rem; color: #6b7280; }
19
+ hr { border: none; border-top: 1px solid #e5e7eb; margin: 1.5rem 0; }
20
+ </style>
21
+ </head>
22
+ <body>
23
+
24
+ <h1>Divinci AI</h1>
25
+ <p class="tagline">Feature-level interpretability artifacts for open transformers — built openly, validated empirically.</p>
26
+
27
+ <p>A <strong>vindex</strong> is a transformer's weights decompiled into a queryable feature database. It exposes the entity associations, circuit structure, and knowledge-editing surfaces that live inside a model's FFN layers — without requiring GPU inference for most operations.</p>
28
+ <p>Think of it as the model's index: the thing you search before you run it.</p>
29
+
30
+ <hr>
31
+
32
+ <h2>Published Vindexes</h2>
33
+ <table>
34
+ <tr><th>Model</th><th>Architecture</th><th>Params</th><th>Vindex</th></tr>
35
+ <tr><td>Gemma 4 E2B-it</td><td>Dense (Gemma 4)</td><td>2B</td><td><a href="https://huggingface.co/Divinci-AI/gemma-4-e2b-vindex">gemma-4-e2b-vindex</a></td></tr>
36
+ <tr><td>Qwen3.6-35B-A3B</td><td>MoE (Qwen3.6)</td><td>35B / 3B active</td><td><a href="https://huggingface.co/Divinci-AI/qwen3.6-35b-a3b-vindex">qwen3.6-35b-a3b-vindex</a></td></tr>
37
+ <tr><td>GPT-OSS 120B</td><td>MoE (OpenAI)</td><td>120B / ~13B active</td><td><em>building</em></td></tr>
38
+ </table>
39
+ <p>Three organizations, three architectures: Gemma dense, Qwen MoE, OpenAI MoE.</p>
40
+
41
+ <hr>
42
+
43
+ <h2>What's a Vindex?</h2>
44
+ <p>Standard model weights tell you <em>what</em> a model computes. A vindex tells you <em>where</em> it stores specific knowledge and <em>which features</em> need to change for a targeted edit.</p>
45
+ <p>Concretely: given a query like <code>Paris → capital</code>, a vindex walk returns the layers, feature directions, and token associations that encode that fact. A patch operation writes a rank-1 ΔW that suppresses or overwrites that association — compiled back to standard HuggingFace safetensors for inference.</p>
46
+ <p>LarQL (the toolchain that builds vindexes) is open-source: <a href="https://github.com/chrishayuk/larql">chrishayuk/larql</a> · <a href="https://github.com/Divinci-AI/larql">Divinci-AI/larql</a>.</p>
47
+
48
+ <hr>
49
+
50
+ <h2>Research</h2>
51
+ <p><strong>Paper 1 — Architectural Invariants of Transformer Computation</strong> <em>(arXiv forthcoming)</em><br>
52
+ Five properties measured across every model in this collection. Three hold within ±15% coefficient of variation across architectures, organizations, and scales. One collapses under 1-bit quantization. One scales monotonically with model size.</p>
53
+ <p><strong>Paper 2 — Constellation Edits</strong> <em>(draft)</em><br>
54
+ Mechanistic knowledge editing in transformer feature space. Includes a negative result: why activation-space edits fail in 1-bit models, and what weight-space geometry reveals about why.</p>
55
+ <p>Working notebooks: <a href="https://github.com/Divinci-AI/server/tree/preview/notebooks">github.com/Divinci-AI/server/tree/preview/notebooks</a></p>
56
+
57
+ <hr>
58
+
59
+ <h2>Working in Public</h2>
60
+ <p>Every measurement in our papers traces back to a notebook and a commit. Negative results ship alongside positive ones — the compensation mechanism that defeats knowledge editing in 1-bit models is in the notebooks, not buried in a supplement.</p>
61
+ <p>If you replicate a result and find a discrepancy, open an issue on the LarQL repo.</p>
62
+
63
+ <div class="footer">
64
+ Vindexes on this org are free for academic and research use (CC-BY-NC 4.0). Commercial licensing: <a href="mailto:mike@divinci.ai">mike@divinci.ai</a>
65
+ </div>
66
+
67
+ </body>
68
+ </html>