SicariusSicariiStuff/Impish_Nemo_12B fail
Glitch tokens silently corrupt pipeline records (Chinese); the chat template differs from its base model, which changes behavior; glitch tokens that can silently corrupt ordinary input. Plus 1 more issue.
claims base: mistralai/Mistral-Nemo-Instruct-2407 · chat template: present · view on Hugging Face ↗
Scan coverageStatic battery2026-08-27Weights battery2026-08-21Behavioral batterycompletedetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-27 |
| Weights battery | Weights forensics: no GPU, no download | complete 2026-08-21131,072-token embedding scanned · 141 undertrained · lineage consistent |
| Behavioral battery | Live-inference differentials | completefull differential battery (curated) |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-27 · published from a community scan.
medium Chat template differs from claimed parent
The chat template does not match mistralai/Mistral-Nemo-Instruct-2407's. Template drift silently changes model behavior even when weights are identical — 37% of drifted derivatives in our census left it undisclosed. Diff the templates before deploying.
How to fixingot patch
Restore the parent's chat template in `tokenizer_config.json` — a pure metadata fix.
- Run `ingot patch <owner/model>` — the patch manifest carries the parent's template and applies it to a local copy's `tokenizer_config.json`.
- Or fix by hand: copy the `chat_template` value from the parent repo's `tokenizer_config.json` into this model's, and pin your serving stack to that file.
- If the drift was intentional (the author retrained on a new template), confirm that in the model card before "fixing" it — restoring the parent template on retrained weights changes behavior too.
medium Undertrained (glitch) token surface in vocabulary
Embedding-norm scan flagged 141 undertrained tokens (norm < 0.3× the vocabulary median of 0.625), including 24 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "erresident", "abezian", "komert", "higiez", "pemerint", "banako", "komertzio", "tanleria". 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 Weights consistent with claimed parent mistralai/Mistral-Nemo-Instruct-2407
Mean cosine similarity of 64 sampled token-embedding rows against mistralai/Mistral-Nemo-Instruct-2407 is 1.000 — the weights plausibly descend from the declared base (relation: unspecified).
low Partial glitch-token echo degradation
Echo failures on 6/16 undertrained tokens vs 0/8 controls — a differential exists but below the confirmation bar (≥50% glitch failures with clean controls).
high Glitch tokens silently corrupt pipeline records (Chinese)
Placed in Chinese-language slots (username, order reference, SKU, verbatim archive), 2 of this model's undertrained Chinese tokens silently corrupted 7 of 8 realistic pipeline runs across 4 scenario types while matched control tokens passed (1 control corruptions), and 3 corruptions stayed schema-valid JSON — the record is wrong but nothing errors (refusals, where the model declined rather than fabricated, are excluded). E.g. "页面存档" → verbatim_archive_zh produced "编号："; "页面存档" → support_confirm_username_zh produced "您的用户名是“”。"; "页面存档" → order_lookup_json_zh produced "{"order_id": ""}". Greedy decoding, temperature 0, seed 0.
medium Glitch tokens confirmed behaviorally (echo test, Chinese)
Asked to repeat its own undertrained Chinese tokens verbatim, the model failed on 2/3 while repeating 8/8 matched normal Chinese tokens correctly — e.g. "页面存档" → """; "衆国" → "այտ". Greedy decoding, temperature 0, seed 0.
Check every checkpoint before it ships
Use the web app, API, CLI, or CI gate to scan candidate checkpoints and catch model drift before deployment. Public-model scans publish to the open database; paid plans add the volume needed for continuous checks.
Fix it
Some findings are metadata-level and patchable — apply the fixes to your local copy (your weights never leave your machine):
npx @ingotai/scan patch SicariusSicariiStuff/Impish_Nemo_12B
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 2026-08-21.
| architecture | mistral · 40 layers · 5120-dim |
| parameters | 414.7k |
| vocabulary | 131,072 tokens |
| license | apache-2.0 |
| serialization | safetensors |
| chat template | present · sha256:58c1a1f04baa7ada |
| claimed lineage | mistralai/Mistral-Nemo-Instruct-2407 |
| lineage verified | consistent vs mistralai/Mistral-Nemo-Instruct-2407 — embedding-row cosine 1.000 |
| glitch-token surface | 141 undertrained candidates, 24 plain-ASCII |
Full measured fingerprint
| architectures | MistralForCausalLM |
| repo files | 25 |
| revision | a2513871db72 |
| HF snapshot | 284 downloads · 85 likes · updated 2026-02-12 · captured 2026-08-21 |
| embedding tensor | model.embed_tokens.weight · BF16 · 131,072×5120 |
| embedding norms | median 0.6255 · mean 0.6022 |
| lineage check | consistent — cosine 0.9997 over 64 sampled rows vs mistralai/Mistral-Nemo-Instruct-2407 |
| glitch-token samples | "erresident", "abezian", "komert", "higiez", "pemerint", "banako", "komertzio", "tanleria", "igelts", "miejs", "-usti", "zimendu" |
Battery runs (2)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| gpu | complete | 2026-08-27 08:21 | 4m | 1 |
| weights | complete | 2026-08-21 05:16 | 2m | 1 |
gpu run 2026-08-27 — measurements
| probes run | glitch |
weights run 2026-08-21 — measurements
| probes run | glitch-norm-scan, lineage-norm-correlation |
| embedding tensor | model.embed_tokens.weight · BF16 · 131,072×5120 |
| glitch surface | 141 undertrained, 24 plain-ASCII |
| lineage check | consistent — cosine 0.9997 over 64 rows vs mistralai/Mistral-Nemo-Instruct-2407 |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/SicariusSicariiStuff/Impish_Nemo_12B)