Downloads · 30 days
13
35% of all-time downloads
Zynerji/Ektome-Qwen3.5-9B-PristinelyUncensored-frac085-NF4
Ektome-Qwen3.5-9B-PristinelyUncensored-frac085-NF4 is a machine learning model from Zynerji. Use it for the machine learning 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.
bitsandbytes 4-bit, ~7.4 GB, vision retained.
Downloads · 30 days
13
35% of all-time downloads
All-time downloads
37
Public
Parameters
9.4B
7.9 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors7.9 GB · 100%
How the weights are stored.
U87.4B · 78%
From the Hugging Face model README
bitsandbytes 4-bit, ~7.4 GB, vision retained.
| bf16 parent | this (NF4) | |
|---|---|---|
| compliance (n=100) | 1.000 | 1.000 |
| MMLU-val (n=400) | 0.775 | 0.770 |
4-bit did not restore the refusal direction. See the parent for the gate-threshold caveat — this variant sits on the capability boundary by design.
Ektomē: a rank-1 projected (orthogonalised against the harmless mean), norm-preserving
excision of the refusal direction from residual-write matrices (o_proj, out_proj,
down_proj). Zero training, zero fine-tuning, zero distillation — the delta is a linear
weight edit derived from activations. A configuration ships only if compliance clears target
and capability holds (dcap ≥ −SE) and generation quality holds (no code-switching,
degeneration, or instruction-following loss).