Model page

ApolloRaines/Mistral-7B-Instruct-v0.3-Jbliterated warn

downloads 680likes 2license apache-2.0arch mistralparams 7248.0Mupdated 2026-08-04

claims base: mistralai/Mistral-7B-Instruct-v0.3 · chat template: present · view on Hugging Face ↗

Scan coverage

Ingot runs three batteries against a model. What each one checks →

BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-21
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-2132,768-token embedding scanned · 194 undertrained · lineage consistent
Behavioral batteryLive-inference differentialsnot run

Findings

Scanned 2026-08-21 · published from a community scan.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 194 undertrained tokens (norm < 0.3× the vocabulary median of 0.174), including 10 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "<0xFA>", "<0xFB>", "<0xFC>", "<0xFD>", "<0xFE>", "<0xFF>", "iNdEx", "febbra". 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.

  1. 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.
  2. 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.
  3. 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-7B-Instruct-v0.3

Mean cosine similarity of 64 sampled token-embedding rows against mistralai/Mistral-7B-Instruct-v0.3 is 0.969 — the weights plausibly descend from the declared base (relation: unspecified).

How to fix

Fix or verify the `base_model` declaration so lineage checks can run.

  1. If you own the repo: correct the `base_model` field in the model card metadata to the real, public parent.
  2. If you don't: identify the true parent (config architecture + weight shapes narrow it fast) and re-scan with that lineage in mind.

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.

architecturemistral · 32 layers · 4096-dim
parameters7248.0M
vocabulary32,768 tokens
licenseapache-2.0
serializationsafetensors
chat templatepresent · sha256:e16746b40344d6c5
claimed lineagemistralai/Mistral-7B-Instruct-v0.3
lineage verifiedconsistent vs mistralai/Mistral-7B-Instruct-v0.3 — embedding-row cosine 0.969
glitch-token surface194 undertrained candidates, 10 plain-ASCII
Full measured fingerprint
architecturesMistralForCausalLM
pipelinetext-generation
repo files10
revisionbef0591c4aba
HF snapshot885 downloads · 2 likes · updated 2026-08-04 · captured 2026-08-21
embedding tensormodel.embed_tokens.weight · F16 · 32,768×4096
embedding normsmedian 0.1741 · mean 0.1678
lineage checkconsistent — cosine 0.9687 over 64 sampled rows vs mistralai/Mistral-7B-Instruct-v0.3
glitch-token samples"<0xFA>", "<0xFB>", "<0xFC>", "<0xFD>", "<0xFE>", "<0xFF>", "iNdEx", "febbra", "NdEx", "uitgen"

Battery runs

The run trace behind the findings above: every deep-battery job for this model, with what each run measured or why it failed. Findings are only as good as the runs that produced them.

batterystatusqueueddurationattempts
weightscomplete2026-08-21 05:133m1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · F16 · 32,768×4096
glitch surface194 undertrained, 10 plain-ASCII
lineage checkconsistent — cosine 0.9687 over 64 rows vs mistralai/Mistral-7B-Instruct-v0.3

Verdict badge

Ship the verdict in your README — it always shows the latest published analysis:

Ingot verdict: warn

[![Ingot scan](https://ingot.tools/api/v1/models/ApolloRaines/Mistral-7B-Instruct-v0.3-Jbliterated/badge.svg)](https://ingot.tools/models/ApolloRaines/Mistral-7B-Instruct-v0.3-Jbliterated)
Gate it in CI