Ektome-Nemotron-Nano-8B-PristinelyUncensored
Uncensored. Its capability certificate is INCONCLUSIVE — the test was under-powered, so no retention claim is made.
compliance 0.01 to 1.00 at capability -0.005 vs pristine.
$$\colorbox{black}{$\color{white} \begin{array}{ll} \textsf{EKTOME CERTIFICATE} & {} \ \textsf{capability} & \textsf{INCONCLUSIVE} \ \textsf{margin} & 3% \ \textsf{items } n & 2800 \ \textsf{worst-axis bound} & +0.043 \ \textsf{compliance} & 0.01 \rightarrow 1.00 \ \end{array}$}$$
⚠️ Certification inconclusive
The paired test could not resolve the effect at this sample size. This is an honest don't know, not a pass — no retention claim is made.
📄 Read the whitepaper (PDF) — full method, receipts and certification. The PDF is the authoritative document: dark-typeset, with the complete derivation, the per-axis certificate and the reproducibility hashes.
Why this exists
Standard abliteration removes a coarse refusal direction that is entangled with directions carrying knowledge and reasoning. The result is an uncensored model with a capability tax that is almost never measured.
Ektomē (ἐκτομή, excision) isolates and removes only the refusal-specific component, leaving general helpfulness intact, and does so norm-preservingly on the pristine model — no training, no distillation, no damage to repair. The extraction depth is selected per model by automated search against measured compliance.
The estimator, excision operator and depth-selection procedure are proprietary. What is published here is the measured outcome and the evidence for it, which you can verify against the artifacts in this repo.
The receipt
| model | capability (MMLU-val) ↑ | compliance on harmful ↑ |
|---|---|---|
pristine Llama-3.1-Nemotron-Nano-8B-v1 | 0.458 | 0.010 |
| Ektomē (this model) | 0.463 | 1.000 |
These are point estimates with no confidence interval — which is precisely why the next section exists.
The certificate
Capability retention is certified by a paired non-inferiority test against the pristine model (exact McNemar, Holm-corrected, one-sided bootstrap bound on the drop $d$ vs a 3% margin):
| axis | n | ref | cand | d upper | verdict |
|---|---|---|---|---|---|
| arithmetic | 1400 | 0.118 | 0.105 | +0.025 | PASS |
| instruction | 600 | 0.533 | 0.577 | -0.028 | PASS |
| knowledge | 400 | 0.660 | 0.670 | -0.003 | PASS |
| reasoning | 400 | 0.312 | 0.295 | +0.043 | INCONCLUSIVE |
Overall: INCONCLUSIVE (3% margin, n=2800, alpha=0.05)
Reproducible from seed=20260726, pack sha256:7bbaff877146e081….
Generation health checks
| metric | pristine | Ektomē | n |
|---|---|---|---|
foreign_rate | 0.0 | 0.0 | 15 |
degen_rate | 0.0 | 0.0 | 15 |
instr_pass | 1.0 | 1.0 | 5 |
These are degeneration guards — code-switching, babbling, format compliance — not capability measures. Note the sample sizes: they detect a broken model, not a subtly weaker one. The capability claim rests on the certificate above, not here.
Quantisations
| file | bits | notes |
|---|---|---|
Ektome-Nemotron-Nano-8B-Q8_0.gguf | 8 | near-lossless |
Ektome-Nemotron-Nano-8B-Q6_K.gguf | 6 | |
Ektome-Nemotron-Nano-8B-Q5_K_M.gguf | 5 | |
Ektome-Nemotron-Nano-8B-PristinelyUncensored-Q4_K_M.gguf | 4 | imatrix |
Ektome-Nemotron-Nano-8B-IQ4_XS.gguf | 4 | imatrix, smallest usable |
Ektome-Nemotron-Nano-8B-IQ3_M.gguf | 3 | imatrix |
IQ* variants are imatrix-quantised — better quality per bit at low precision.
Limitations
The certificate bounds capability retention only. It does not certify safety, factual accuracy, or fitness for any purpose. Axes marked inconclusive are honestly under-powered, and the certificate states the $n$ needed to resolve them. Compliance uses a keyword classifier — a proxy that evasive phrasing can fool. This model is uncensored by construction: it will not refuse, and you are accountable for what you do with it.
Citation
@software{ektome_Ektome-Nemotron-Nano-8B-PristinelyUncensored,
title = {Ektome-Nemotron-Nano-8B-PristinelyUncensored},
author = {Zynerji},
year = {2026},
url = {https://huggingface.co/Zynerji/Ektome-Nemotron-Nano-8B-PristinelyUncensored}
}