SOFTWARE / ADVISORY KERNEL · SURROGATE WEIGHTS UNAVAILABLE · NO PROMOTION

This model-type repository is a legacy source/card mirror for the weighted-geometric-mean kernel. The receipt-named model.safetensors and config.json are absent. MODEL_PROVENANCE.json records trained_weights_present: false; TRAINING_RECEIPT.json retains historical surrogate metrics. OPERATIONAL.json and BENCH.laptop-blackwell.json retain the documented compile failures. No new surrogate release, deployment readiness, model promotion, or autonomous authority follows from this card.

Λ uniqueness remains Conjecture 1: OPEN and advisory. A signature or selfcheck does not establish proven trust. The predecessor and compatibility relationship is szl-governed-norm; canonical kernel source is szl-lambda-gate. Kernel Hub migration references below describe a separate repository type.

CORRECTION 2026-08-30 — the surrogate weights are NOT in this repo

A callout below stated that this repository "also ships" model.safetensors, config.json with MEASURED fidelity 0.9670. They are not here — every one returns HTTP 404 on resolve/main. MODEL_PROVENANCE.json also asserted trained_weights_present: true with a sha256 for the missing file; that attestation has been corrected in the same commit and now reads false, with the digest retained as the expected value for when the weights are pushed.

What remains true: the kernel is real, get_kernel is import-LIVE, and TRAINING_RECEIPT.json documents a genuine training run, so the fidelity figures keep their provenance. What was false: the claim that the resulting artifact is downloadable from this repo. Do not build against the surrogate here — there is nothing to load. The kernel was always the declared ground truth; that part of the card was correct and is unchanged.

Operational (MEASURED laptop-Blackwell)

STATUS: tests FAIL. get_kernel import-LIVE. Unsloth/LoRA is the wrong tool. Receipted kernels, not silent CUDA.

Thing	Label	Method / N / date / what-NOT
tests (PYTHONPATH=torch-ext)	FAIL	MEASURED 2026-08-29T15:53:47Z host betterwithage Windows-10-10.0.26200-SP0. torch 2.10.0+cu128. GPU NVIDIA GeForce RTX 5050 Laptop GPU arch Blackwell. pytest 4 failed, 54 passed, 14 warnings in 52.57s. Failed nodes: tests/test_lambda.py::test_torch_compile_friendly; tests/test_lambda.py::test_fullgraph_lambda_aggregate; tests/test_lambda.py::test_fullgraph_lambda_gate_score; tests/test_lambda.py::test_fullgraph_lambda_gate_batch_score. What-NOT: not a leaderboard. torch.compile fullgraph failures on Windows Blackwell (cl is not found) are MEASURED, not hidden.
Kernel Hub get_kernel	import-LIVE	kernels 0.16.1. Default: get_kernel("SZLHOLDINGS/szl-lambda-gate", revision="main", trust_remote_code=True) → True. backend="cpu" → True. trust_remote_code=False → ValueError (SZLHOLDINGS is not a trusted publisher). repo_type=kernel required (kernels 0.16). What-NOT: not a weight load; do not pickle/joblib.load.
formula-tax	ADVISORY	locked-8 F1 F4 F7 F11 F12 F18 F19 F22. registry_count=21. Λ geomean 0.316227766016838. uniqueness Conjecture 1 (never a theorem).
I1–I8	catalog	I1 receipt-chain-continuity; I2 ledger-failure-shape; I3 served-run-has-model; I4 signed-columns-atomic; I5 loop-steps-positive; I6 receipt-ed25519-verify; I7 receipt-columns-consistent; I8 flywheel-lineage. Executed by SZLHOLDINGS/szl-invariants. Statuses never coerced. Λ untouched.
CUDA speedup / tokens/s / joules	UNAVAILABLE	Not claimed. Receipted kernels, not silent CUDA.

GitHub source: szl-holdings/szl-lambda-gate @ 3fb5bb65dfdba006ae917e10b7703a535ea303e0. Artifacts: BENCH.laptop-blackwell.json, OPERATIONAL.json.

from kernels import get_kernel
k = get_kernel("SZLHOLDINGS/szl-lambda-gate", revision="main", trust_remote_code=True)

Source-only CPU follow-up — 2026-09-24

An owner-run local check tested GitHub source f7e290a197922134076ff190118d7fca2e4975db on Windows 10 build 26200, Python 3.11.9, CPU PyTorch 2.14.0, and MSVC 14.44.35207. No GPU was used. The unmodified source suite passed 62 tests, with no failures or skips, including all four compilation tests, after activating the installed MSVC developer environment. A separate explicit run of the folded normalization tests passed 110 tests, with no failures or skips. Those governed_norm_test_*.py files are not collected by default pytest discovery. Kernel code and test thresholds were unchanged.

The source-only test record names the exact source, scope, run times, and retained log/JUnit files with their hashes. It is unsigned owner-run evidence. The dated operational failures above remain unchanged. These new results do not qualify the current Hub model build, the separate first-class Kernel Hub repository, CUDA execution, production deployment, or autonomous authority. Separate Hub-build probes were incomplete after disk-space exhaustion and provide no runtime pass or diagnosed failure.

_{Part of the SZL Holdings governed estate — claims are designed to carry checkable receipts. Verification proves integrity & origin, never accuracy or performance.}

🟥 Kernel real; surrogate weights NOT PUBLISHED — see the correction at the top. The Λ governance kernel (pure-torch, differentiable) is UNCHANGED and remains the sole ground truth. Since surrogate v1 this repo was described as shipping (IT DOES NOT — 404) model.safetensors + config.json — a real trained tiny torch MLP that predicts the ADVISORY gate decision lambda_gate(axes, threshold).passed over the 13-axis Yuyay space, with MEASURED fidelity 0.9670 (agreement vs the kernel on a held-out split). The surrogate approximates the gate DECISION only; the kernel Λ stays authoritative and get_kernel-discoverable. Λ is the weighted geometric mean, NOT proven trust — uniqueness = Conjecture 1 (OPEN).

szl-lambda-gate

Kernel Hub migration (verified 2026-07-15): get_kernel(...) now resolves the matching first-class Kernel Hub repository. Its main and stable v1 refs both pin verified revision 47c7eb2db8859507d4115adbbff20c65de66dbb5. This model-type repository is retained as the legacy source/card mirror.

Λ — a governance aggregator as a Hugging Face kernel. A differentiable weighted-geometric-mean aggregator with an ADVISORY non-compensatory gate and runtime axiom self-checks, from SZL Holdings. Its torch.compile support is limited by the documented failures above.

Companion to szl-governed-norm. Where that kernel makes a normalization auditable, this one makes a governance decision computable and checkable at the tensor layer.

Interactive demo

The cut

A gate you can import in PyTorch. Fail-closed by default.

The smallest possible governed op.

Silhouette → leave → SZL

Leader	Take, then tweak
Anthropic	Refuse compiled to autograd.
NVIDIA	Custom op, NVIDIA-shaped packaging.
Unsloth	No.

Nobody else ships this combination. That is the point of a one-of-one.

Intended use

Torch forward-pass gate.

Limitations

• Surrogate. Not Conjecture-1 solved.

Canonical GitHub: szl-holdings/szl-khipu

Live demos (in-browser, nothing to install) — lambda-gate-holo (this kernel's holographic gate demo) · szl-kernels-live (unified suite demo).

The quickstart above runs fully locally. For a full governed-kernel suite demo, see szl-kernels. For the live a11oy substrate, see a11oy Space.

What Λ is — and is NOT (read this first)

Λ is the weighted geometric mean over axis scores in [0,1]:

[ \Lambda(x) = \prod_i x_i^{w_i}, \quad \sum_i w_i = 1, ; w_i > 0, ; x_i \in [0,1] ]

It is a non-compensatory, ADVISORY roll-up: any single zeroed (or non-finite) axis drives the whole aggregate to 0 — a conservative "one bad axis fails the gate" signal. Λ is NOT "proven trust" and NOT a closed theorem. Its uniqueness (that the weighted geometric mean is the only aggregator satisfying the carried axioms) remains Conjecture 1 — OPEN. A gate "pass" is an advisory signal, never a guarantee. We label this honestly everywhere.

Quickstart

pip install kernels torch

import torch
from kernels import get_kernel

# Current `kernels` (>=0.15) requires an explicit revision/version + trust flag for org kernels:
lg = get_kernel("SZLHOLDINGS/szl-lambda-gate", revision="main", trust_remote_code=True)
# (once a tag is published you can pin it, e.g. revision="v0.2.0")

axes = torch.tensor([0.9, 0.8, 0.95])      # axis scores in [0,1]
score = lg.lambda_aggregate(axes)          # Λ(x) ∈ [0,1]
res   = lg.lambda_gate(axes, threshold=0.5)
print(res.score, res.passed, res.advisory) # advisory is always True

print(lg.selfcheck())                      # empirical A1–A4 checks + version

API

Function	Notes
lambda_aggregate(axes, weights=None)	Λ over the last dim. Differentiable, batched, torch.compile-friendly.
lambda_gate(axes, weights=None, threshold=0.5)	Advisory gate → LambdaGateResult(score, passed, threshold, advisory).
lambda_gate_batch(candidates, weights=None, threshold=0.5)	Score many candidate vectors (..., N, k) in one call; returns the advisory pass mask.
selfcheck()	Empirical A1–A4 axiom checks + adversarial falsification search + version. NOT a uniqueness proof.
is_monotone / is_homogeneous / is_egyptian_exact / is_bounded_by_max	The four carried axioms as real runtime checks.
yuyay_weights(), YUYAY_AXES, YUYAY_FLOORS	Canonical 13-axis Yuyay preset (advisory).
layers: LambdaGate, LambdaAggregate	Pure nn.Module for the Kernel Hub layer-mapping mechanism.

Carried axioms (verifiable, not a proof)

• A1 IsMonotone — Λ is non-decreasing in each axis.
• A2 IsHomogeneous (deg 1) — Λ(t·x) = t·Λ(x).
• A3 IsEgyptianExact — Λ(c,…,c) = c.
• A4 IsBounded — Λ(x) ≤ maxᵢ xᵢ.

selfcheck() verifies these empirically on sampled inputs and runs a random falsification search. A clean run is evidence, not proof — Λ-uniqueness is Conjecture 1 (open).

Provenance

Backed by the Lean 4 formalization szl-holdings/lutar-lean (749 declarations / 14 axioms / 163 tracked sorries), DOI 10.5281/zenodo.20434308. Λ uniqueness = Conjecture 1 (open).

Honesty

• Pure-Python universal kernel — a correctness reference, not a CUDA speed record. No fabricated benchmarks (50 passing tests).
• Λ is advisory; never "proven trust."
• Prior art honestly attributed: the weighted geometric mean as a less-compensatory composite indicator is established practice (UN HDI 2010, OECD Composite Indicators Handbook 2008); the veto/cut-off idea is ELECTRE. The 13-axis conjunctive form is SZL's own yuyay_v3 gate.

Compatibility

Python 3.9+, torch>=2.5, standard library + torch only.

License

Apache-2.0. Copyright 2026 SZL Holdings.

Historical Λ-gate surrogate v1 (receipt-reported metrics; weights unavailable)

TRAINING_RECEIPT.json reports a tiny torch MLP (3 hidden ReLU layers, 64 units; the absent model.safetensors + config.json) trained on 40,000 axis-score vectors synthesized and labeled by this kernel itself (lambda_gate(axes, weights=yuyay_uniform_1/13, threshold=0.5).passed, seed 20260721; 800 samples re-audited by independent full kernel replay during generation — all agreed). Inputs are the 13 Yuyay axis scores in [0,1], including non-compensatory zero-route rows (a single zeroed axis must fail the gate).

metric	value
fidelity vs kernel (held-out agreement)	0.9670
recall GATE_PASS	0.9912
recall GATE_FAIL	0.9469

Honest boundary: the surrogate learns the decision boundary of an ADVISORY, non-compensatory aggregator; it is a fast approximation, NOT the exact Λ and NOT proven trust. Residual disagreement lives near the Λ=threshold surface — the exact kernel lambda_gate remains authoritative. Class counts: GATE_FAIL=21828, GATE_PASS=18172. Λ uniqueness = Conjecture 1 (open).

Historical evidence only — surrogate loading is unavailable. TRAINING_RECEIPT.json reports the metrics above and names model.safetensors plus config.json; neither file is present in this repository. MODEL_PROVENANCE.json preserves the expected weight digest without attesting that the bytes are published. Do not attempt to load either missing file. At the revision inspected on 2026-09-24, claims.trained_model=SURROGATE_PRESENT_MEASURED_FIDELITY contradicted the false presence fields. The 2026-09-24 metadata correction changes only that label to SURROGATE_UNAVAILABLE_RECEIPT_REPORTED_FIDELITY. Correction-file SHA-256: 731bd283b2ee7ae0732b8468b7c13e56e87d713a35d3e8a4d5e7e9d9cdeb49fb. The immutable original record and historical metrics remain available; this correction does not publish or evaluate surrogate weights. These historical surrogate metrics cannot be replayed from this repository until the named artifacts are published and the weight hash is verified. The retained scripts/eval.py is historical verification tooling, not a working surrogate replay path for the current file set. The exact kernel remains the reference for the advisory decision; the documented compile failures remain open.

---

SZL Kernels Suite

Part of the szl-kernels governed-kernel suite — the hub links every member, and each member links back to the hub so no leaf is orphaned:

Kernel	Lane
szl-kernels	hub — unified suite, cross-kernel UnifiedReceiptChain
szl-governed-norm	RMSNorm/LayerNorm + SHA3-256 receipts
szl-lambda-gate (this repo)	advisory Λ gate (Conjecture 1, OPEN)
governed-inference-meter	MEASURED-joule energy accounting (NVML)
szl-govsign	signed governance attestation (DSSE / in-toto)
szl-blocked	honest-BLOCKED state + EU AI Act Annex IV DRAFT
szl-provctl	provenance-DAG verify + in-toto/SLSA interop

Live Spaces: a11oy · hatun-mcp.

Related — Governed Kernels collection: Governed Kernels & Verifiers groups the whole family in one page. Live console: a11oy · a-11-oy.com · llm-router · receipt verifier · receipt spec (hub).

---

_{SZL Holdings · Λ governance aggregator · advisory, not proven trust · a-11-oy.com · github.com/szl-holdings · huggingface.co/SZLHOLDINGS}

---

Citation

Cite this. Part of the SZL Holdings Ouroboros Thesis (Governed Post-Determinism).
Concept DOI (always-latest): 10.5281/zenodo.19944926.
Author: Stephen P. Lutar Jr. · ORCID 0009-0001-0110-4173 · License CC-BY-4.0.
Full DOI-pinned lineage (v1→v26) + the 8 papers: szl-papers PAPERS_INDEX.
No artifact-specific DOI is minted for this model; the concept DOI above covers the program.

Honesty (Doctrine v11): Λ unconditional uniqueness is Conjecture 1 (machine-checked FALSE as stated) — never a theorem; conditional uniqueness is Theorem U (axiom-free). Locked-proven formulas = exactly 8 {F1,F4,F7,F11,F12,F18,F19,F22}; ~185 experimental theorems are a separate CI-green tier; Khipu BFT safety = Conjecture 2. Trust never 100%.

@misc{lutar_szl_ouroboros,
  author    = {Lutar, Stephen P., Jr.},
  title     = {SZL Holdings --- The Ouroboros Thesis (Governed Post-Determinism)},
  year      = {2026},
  publisher = {Zenodo},
  doi       = {10.5281/zenodo.19944926},
  url       = {https://doi.org/10.5281/zenodo.19944926},
  note      = {Concept DOI --- always resolves to the latest version. ORCID 0009-0001-0110-4173. CC-BY-4.0.}
}

Signed-off-by: Stephen Lutar stephenlutar2@gmail.com

Files in this repo

Path	What it is
build/torch-universal/szl_lambda_gate/__init__.py	public API — lambda_aggregate, lambda_gate, selfcheck()
build/torch-universal/szl_lambda_gate/_lambda.py	the weighted-geometric-mean aggregator + A1–A4 empirical checks
build/torch-universal/szl_lambda_gate/layers.py	nn.Module wrapper
build.toml · build/torch-universal/szl_lambda_gate/metadata.json	Kernel Hub build/metadata manifests
LICENSE · SECURITY.md	Apache-2.0 · security policy

---

SZL Holdings · a-11-oy.com · szl-kernels

_{SLSA: L1 honest · L2 attested · L3 roadmap. Λ = Conjecture 1 (advisory, never a theorem). Trust ceiling 0.97 — never 100%. Labels honest by default: MEASURED / REPORTED / MODELED / HEURISTIC / UNKNOWN / UNAVAILABLE. locked-proven = exactly 8 {F1,F4,F7,F11,F12,F18,F19,F22}.}

Estate wiring

Source | Estate hologram | Receipt ledger | Product | Proof

Verification proves integrity and declared origin only; it does not prove accuracy, readiness, or performance.