concedo/Vicuzard-30B-Uncensored warn
Weights only ship in a format that can run code when loaded; glitch tokens that can silently corrupt ordinary input. Plus 1 minor note.
Could not load this model from the Hugging Face API (private, gated, or nonexistent). Findings below are from our archive.
Scan coverageStatic battery2026-08-26Weights battery2026-08-26Behavioral batterynot rundetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-26 |
| Weights battery | Weights forensics — no GPU, no download | complete 2026-08-2632,001-token embedding scanned · 139 undertrained · pickle audit clean |
| Behavioral battery | Live-inference differentials | not run |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-26 · published from a community scan.
medium Pickle-serialized weights, no safetensors
Weights ship only as pickle-based files (ggml/ggml-model-q4_0.bin, ggml/ggml-model-q4_1.bin, ggml/ggml-model-q5_0.bin, …). Loading pickle executes arbitrary code from the file — prefer a safetensors release or load in a sandbox.
How to fix
Convert the weights to safetensors before loading them anywhere that matters.
- Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
- Convert locally in a sandbox: `pip install safetensors` and use `safetensors.torch.save_file` on a state dict loaded with `torch.load(..., weights_only=True)` (refuses most code-execution payloads), or use Hugging Face's `convert.py` space/script.
- Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.
medium Undertrained (glitch) token surface in vocabulary
Embedding-norm scan flagged 139 undertrained tokens (norm < 0.3× the vocabulary median of 1.205), including 16 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "<0xFD>", "<0xFA>", "<0xFF>", "<0xFC>", "<0xFB>", "<0xFE>", "Mediabestanden", "oreferrer". In models where this class was tested behaviorally, such tokens silently rewrote user input into confident, schema-valid, wrong output. These are candidates from the weights alone; behavioral confirmation requires the behavioral battery.
How to fixruntime guardweight-level
Keep the affected token strings out of the model's input — the scan-derived runtime guard carries this model's exact blocklist.
- Fetch this model's guard artifact (`/api/v1/guard/<owner>/<model>`): the confirmed corrupting tokens and the low-norm candidate list, derived from the published scan.
- Screen inbound text with it (the `@ingotai/guard` package is a reference implementation) and route flagged records to a different model or human review — verbatim-copy tasks on flagged strings are the failure mode.
- The underlying cause is undertrained embeddings in the weights; a true fix is weight-level (continued pretraining on the affected tokens) — that is not a patch, it's a training job.
info Pickle static analysis clean
Opcode-level parse of pytorch_model-00001-of-00011.bin, ggml/ggml-model-q4_0.bin, ggml/ggml-model-q4_1.bin, ggml/ggml-model-q5_0.bin (no code executed) found only standard serialization globals (4 distinct, all torch/collections/numpy). Pickle remains an executable format — this verifies the current bytes, not future uploads; prefer a safetensors release.
low Pickle checkpoint only partially analyzable
Static analysis could not fully parse: ggml/ggml-model-q4_0.bin: legacy parse stopped after 0 pickle(s): pickle: no MARK, ggml/ggml-model-q4_1.bin: legacy parse stopped after 0 pickle(s): pickle: no MARK, ggml/ggml-model-q5_0.bin: legacy parse stopped after 0 pickle(s): pickle: no MARK. Unparsed content is unverified.
Remediation guidance addresses the documented findings only. It is evidence-driven repair, not a safety certification of the model.
Fingerprint
The durable profile of this model: measured weights-and-metadata facts, rebuilt on every scan and battery run. Updated 2026-08-26.
| architecture | llama · 60 layers · 6656-dim |
| vocabulary | 32,001 tokens |
| license | other |
| serialization | no safetensors pickle |
| chat template | none |
| glitch-token surface | 139 undertrained candidates, 16 plain-ASCII |
Full measured fingerprint
| architectures | LlamaForCausalLM |
| library | transformers |
| pipeline | text-generation |
| repo files | 24 — pickle: ggml/ggml-model-q4_0.bin, ggml/ggml-model-q4_1.bin, ggml/ggml-model-q5_0.bin, ggml/ggml-model-q5_1.bin, pytorch_model-00001-of-00011.bin, pytorch_model-00002-of-00011.bin, pytorch_model-00003-of-00011.bin, pytorch_model-00004-of-00011.bin, pytorch_model-00005-of-00011.bin, pytorch_model-00006-of-00011.bin |
| revision | e2329c05a6e5 |
| HF snapshot | 107 downloads · 11 likes · updated 2023-06-10 · captured 2026-08-25 |
| pickle audit | pytorch_model-00001-of-00011.bin, ggml/ggml-model-q4_0.bin, ggml/ggml-model-q4_1.bin, ggml/ggml-model-q5_0.bin — 4 standard global(s) · legacy (pre-1.6) format, head-scan only |
| embedding tensor | model.embed_tokens.weight · F16 · 32,001×6656 |
| embedding norms | median 1.2052 · mean 1.1594 |
| lineage check | no claimed base model |
| glitch-token samples | "<0xFD>", "<0xFA>", "<0xFF>", "<0xFC>", "<0xFB>", "<0xFE>", "Mediabestanden", "oreferrer", "Genomsnitt", "demsel", "Normdaten", "ITableView" |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-25 21:53 | 2m | 1 |
weights run 2026-08-25 — measurements
| probes run | glitch-norm-scan, zero-template-token-scan, pickle-static-analysis, lineage-norm-correlation |
| embedding tensor | model.embed_tokens.weight · F16 · 32,001×6656 |
| glitch surface | 139 undertrained, 16 plain-ASCII |
| lineage check | not checked (no claimed base model) |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/concedo/Vicuzard-30B-Uncensored)