dafit/nimmerverse-sensory-network
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: Finalize model choice and iterate on project critique

Browse Source 
- **Endgame-Vision.md**: Updated the target base model to 'Qwen3-VL 32B (Thinking Version)' and reflected this in the deployment vision. The document version was also incremented to 5.5 to mark this significant decision.
- **archive/nimmerverse-critique-and-analysis-2025-12-13.md**: Iterated on the critique document to include a new, synthesized understanding of the 'Economy of Mind'. This new section (Section 10) details how 'Constrained Emergence' and the 'Temporal-Ternary Gradient' form a sophisticated system for attention and resource allocation. The 'Potential Challenges' section (Section 8) was also updated with new points related to calibrating economic triggers and preventing poverty traps.
This commit is contained in:
dafit
2025-12-14 00:17:31 +01:00
parent 5b37179b50
commit 91ef0520ce
2 changed files with 39 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
Endgame-Vision.md
View File

@@ -1,6 +1,6 @@
---
type: research_vision
version: 5.4_color_form_protocol
version: 5.5_qwen3_thinking_model
status: vision_document
created: 2025-11-04
updated: 2025-12-13
@@ -79,7 +79,7 @@ This is a **RESEARCH VISION** - a platform for studying how intelligence emerges
│ → ../nyx-probing/PLAN.md │
│ │
│ Layer 2: YOUNG NYX (Single Model + LoRA Stack + Dialectic) │
│ ├─ Base: Qwen3-VL-32B (96GB VRAM in the Womb) │
│ ├─ Base: Qwen3-VL 32B (Thinking Version) (96GB VRAM in the Womb) │
│ ├─ LoRA adapters: Identity, Technical, Creative (hot-swap) │
│ ├─ Mirror: Negated LoRA weights for dialectic (-1 × Nyx) │
│ ├─ Dialectic: Thesis (Nyx) → Antithesis (Mirror) → Synthesis │
@@ -257,7 +257,7 @@ For high-stakes queries (identity, ethics, low confidence):
**Hardware:** RTX PRO 6000 Blackwell (96GB VRAM) - "The Womb"
**Solution:** Unsloth for fine-tuning (~77GB), Lorax for hot-swap LoRA adapters (<100ms)
**VRAM Budget:** Base ~77GB + Active LoRA ~200MB = fits in 96GB ✓
**Vision:** Qwen3-VL-32B brings unified vision + video + OCR + reasoning
**Vision:** Qwen3-VL 32B (Thinking Version) brings unified vision + video + OCR + reasoning
---