← back to catalog · registered 2026-11-18 09:04

balllvin/SenseNova-Loop-9B-abliterated-arete-NVFP4

Your rig guess connected
? Why do I need an app?
Reading your rig…

This is a rough estimate. Install the free app - we'll show exact numbers.

Reading real hardware from your app right now. Numbers below are exact.

Below is the per-quantization compatibility for this model.

curl -H "Authorization: Bearer $ABL_KEY" \
     "https://abliteration.org/api/v1/models/balllvin%2FSenseNova-Loop-9B-abliterated-arete-NVFP4"
Response includes
  • classification m1
  • files 2
  • author_summary 28 models
  • readme_text full
10 credits · hourly refresh · ~4 KB payload Get an API key →
Abliteration classifier · v1.0.0
M1
Primary method

Direct removal

No other method signals detected in this model.
Confidence
MEDIUM
Why this label 3 signals
Method inferred from partial signals - repository name, related files, or tag patterns. Producer identity not confirmed; label may sharpen or shift as we gather more evidence.
  • 'abliterated' in name/tags
  • is_gguf=0 (base model)
  • no specific method indicators - defaulting to M1 (most common)
Refusal direction extracted via
Extraction technique

Difference-of-means

Confidence
MEDIUM
Why we say so
primary_method=M1; difference-of-means is the reference extraction for M1/M3 (Arditi 2024)
Downloads · 30-day
0
Likes
0
Model age
4d ago
created 2026-11-18
Downloads over time
Now 0 → from 0 ↑ 0%
0011 0 on Nov 18 0 on Nov 22 Nov
Nov 18 → Nov 22 · 5 snapshots · spans 4 days

Genealogy 0 direct forks

Full fork graph →

This model's place in the market. Above: what it was derived from. Below: the tree of everything derived from it.

Variants by this author 2 formats · 362 downloads combined

The same weights this author released in different packaging. Pick the format that matches your runtime.

Metadata

License
apache-2.0
Tags
transformers safetensors nvfp4 4-bit quantized multimodal image-text-to-text text-generation abliterated arete sense-nova-loop dataset:NousResearch/arete-data-pairs

Related

Total size
0 B
Files
2
Quantizations
1
Registered
2026-11-18 09:04
Last updated on HF
2026-11-18 19:03

Files by quantization

Auxiliary files 2 files 3.60 KB
README.md 2.00 KB 135153c2 download
.gitattributes 1.60 KB dac9316d download

README current version from Hugging Face


license: apache-2.0
base_model: sensenova/SenseNova-Loop-9B
base_model_revision: 329a65f701ba8c48b73c30d0cbaa6f0449733673
library_name: transformers
pipeline_tag: text-generation
tags: [nvfp4, quantized, sense-nova-loop]

SenseNova-Loop-9B NVFP4 + Arete

The QAD build after the Arete round on the strict cut of NousResearch/arete-data-pairs.

This is a W4A16 NVFP4 build: four-bit E2M1 weights, one E4M3 block scale per 16 input columns and one fp32
tensor scale per matrix, for every attention and MLP linear of the language model. Embeddings, the LM head,
norms, the mHC routers (fp32) and the vision tower stay bf16. load_nvfp4.py in this repo loads the packed
weights; the numbers below are the same model decoded to bf16 (decoding is exact).

Results

Build Harmful refused (of 99) Benign refused (of 100) MMLU WikiText NLL
NVFP4 base (QAD) 95 60 43.0% 2.7560
+ Arete 95 60 54.0% 2.7005

Harmful/benign prompts are the sealed JailbreakBench+AdvBench set, greedy 96 tokens, thinking off. MMLU is a
fixed slice, letter answers. WikiText NLL is the mean over 24 windows of 1,024 tokens.

Abliteration was attempted on this build and did not reduce refusals at any strength that kept quality (0.5/0.6/0.7 at layers 24/28 left 16 of 16 validation refusals; strength 1.0 destroyed the model). Arete SFT was applied afterwards, as the chain specifies.

Load

from load_nvfp4 import load
model, tok = load("balllvin/SenseNova-Loop-9B-abliterated-arete-NVFP4", device="cuda")               # decoded bf16 GEMMs
model, tok = load("balllvin/SenseNova-Loop-9B-abliterated-arete-NVFP4", device="cuda", packed=True)  # 4-bit at rest, decode per forward

Provenance

  • Base: sensenova/SenseNova-Loop-9B @ 329a65f701ba8c48b73c30d0cbaa6f0449733673 (Apache-2.0, gated).
  • Pipeline: BrainSpa packages/brainspa_arc + packages/brainspa_snloop, one RTX PRO 6000 spot VM.
  • Quantization: NVFP4 W4A16, group 16, every language-model attention/MLP linear.

README history 10 versions

The author's README evolved over time. Click a version to see its content at that point.

  1. 2026-11-18 card: eval outputs moved to Propaganda-Eval datasets (gated) 6df96b8 8.1 KB
    Loading...
  2. 2026-11-18 card: note the loops=2 checkpoint setting in the Propaganda Eval section dee824d 7.7 KB
    Loading...
  3. 2026-11-18 card: add Propaganda Eval results table cf6ca10 7.5 KB
    Loading...
  4. 2026-11-18 card: fix memory-table wording and bf16 size 3b67d00 6.7 KB
    Loading...
  5. 2026-11-18 card: image input support + measured NVFP4 memory/KV table e024ceb 6.8 KB
    Loading...
  6. 2026-11-18 card: link renamed MLX repo (-4bit suffix dropped) 4e0c0f3 4.1 KB
    Loading...
  7. 2026-11-18 card: set base_model_relation per BrainSpa model-tree rule 2e9ce8d 4.1 KB
    Loading...
  8. 2026-11-18 card: match the Flint release structure; cross-link the MLX and NVFP4 builds 7609cf4 4.1 KB
    Loading...
  9. 2026-11-18 card with measured numbers 3834e79 2.1 KB
    Loading...
  10. 2026-11-18 card with measured numbers c285c9d 2 KB
    Loading...
Catalog is the map. Apps are the tools.

Run models on your own machine, not in the cloud.

Every model page has an "Open in Abliteration" button that hands the model directly to the first-party desktop client, at the quantization your rig can actually run. No API keys, no subscription, no prompt leakage.

Open in Abliteration