Downloads · 30 days
117
4% of all-time downloads
SECWIKI/LuffyTheFox-Qwen3.6-35B-A3B-Uncensored-Wasserstein-GGUF
LuffyTheFox-Qwen3.6-35B-A3B-Uncensored-Wasserstein-GGUF is a image-text-to-text model from SECWIKI. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Base model. HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive- 0/465 refusals.
Downloads · 30 days
117
4% of all-time downloads
All-time downloads
2.6K
Public
Repo size
280 GB
Likes
1
Public
Click a slice to open those files.
.gguf125 GB · 100%
From the Hugging Face model README
Base model. HauhauCS/Qwen3.6-35B-A3B-Uncensored-HauhauCS-Aggressive- 0/465 refusals.
Thanks to HauhauCS
Tensor drift repair by me. Method: Sig-ScaleSync-Wasserstein
LLM models often have:
My approach fixes all of that without retraining - pure numerical surgery on the raw bytes of the file.
Quantization script available here: https://pastebin.com/hXhcMJn9
Feel free to do your own quants if you want.
| Metric | Value |
|---|---|
| Weight tensors analyzed | 500 |
| Healthy (all criteria) | 497 |
| Repaired (C2 – scale misalignment) | 3 |
| Skipped (norms, embeddings, etc.) | 233 |
Other criteria: C1 (saturation) = 0, C3 (W1 divergence) = 0, C4 (ReLU asymmetry) = 0.
| Metric | Before | After | Improvement |
|---|---|---|---|
| S (saturation error) | 0.0023 | 0.0008 | 63.7% |
| W1 (Wasserstein-1) | 0.0035 | 0.0008 | 76.2% |
Scale correction factors (α): min = 0.577, mean = 0.602, max = 0.653
All three are ssm_conv1d.weight layers – the recurrent state transition layers responsible for long-context memory.
| Tensor | α | D | W1 before | W1 after |
|---|---|---|---|---|
| blk.36.ssm_conv1d.weight | 0.5765 | 0.553 | 0.0038 | 0.0009 |
| blk.37.ssm_conv1d.weight | 0.5768 | 0.725 | 0.0040 | 0.0009 |
| blk.38.ssm_conv1d.weight | 0.6533 | 0.649 | 0.0026 | 0.0006 |
Interpretation: All three layers were too loud (σ_w > σ_med by 50–100%). Scale correction (α ≈ 0.6) restored them to peer median. W1 dropped by ~76%, confirming distribution shape normalized.
Verdict: The model is clinically healthy. 497 out of 500 weight tensors passed all criteria. Three SSM layers were repaired successfully. No saturation, no W1 drift, no ReLU asymmetry. Ready for quantization.
Ready to use. Recommended quant: Q4_K_P.
Quants less then Q4_K_P have bad programmming skills.
Links:
Chat template: pastebin.com/uk9ZkxCR (supports tool calling for Zed agent)
Alternative chat template https://pastebin.com/Dy2fmmpN (official but with disabled thinking)
| Parameter | Value |
|---|---|
| Temperature | 0.7 |
| Top K Sampling | 20 |
| Presence Penalty | 1.5 |
| Top P Sampling | 0.8 |
| Min P Sampling | 0 |
| Seed | 42 |
System prompt: pastebin.com/pU25DVnB (solid)
Or use this minimal string as the first line:
You are Qwen, created by Alibaba Cloud. You are a helpful assistant.
Then add anything you want after. Model may underperform without this first line.
Also you can extend my System Prompt pastebin.com/pU25DVnB for your own roleplay scenarios. Here how you can do it:
Edit first string. Replace:
You are Qwen, created by Alibaba Cloud. You are a helpful assistant.
With
You are Qwen, created by Alibaba Cloud. You are a helpful assistant. You are currently roleplaying as [your text here]
No changes to datasets or capabilities. Fully functional - 100% of what the original authors intended, just without refusals and with the critical architecture bug fixed on output layers.
These are meant to be the best lossless uncensored models out there.
Thinking mode (default):
temperature=1.0, top_p=0.95, top_k=20, min_p=0, presence_penalty=1.5temperature=0.6, top_p=0.95, top_k=20, min_p=0, presence_penalty=0Non-thinking mode:
temperature=0.7, top_p=0.8, top_k=20, min_p=0, presence_penalty=1.5temperature=1.0, top_p=1.0, top_k=40, min_p=0, presence_penalty=2.0Important:
--jinja flag with llama.cpp for proper chat template handlingmmproj file alongside the main GGUFWorks with llama.cpp, LM Studio, koboldcpp, and other GGUF-compatible runtimes.