Model page

KeefeBuild/Keefe-Discere warn

The chat template differs from its base model, which changes behavior; glitch tokens that can silently corrupt ordinary input. Plus 1 minor note.

downloads 1.7klikes 1license apache-2.0arch qwen2params 7615.6Mupdated 2026-08-23

claims base: Qwen/Qwen2.5-7B-Instruct, Qwen/Qwen2.5-Coder-7B-Instruct, Qwen/Qwen2.5-Math-7B-Instruct · chat template: present · view on Hugging Face ↗

Scan coverageStatic battery2026-08-21Weights battery2026-08-21Behavioral batterynot rundetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-21
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-21152,064-token embedding scanned · 7806 undertrained · lineage inconclusive
Behavioral batteryLive-inference differentialsnot run

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

Findings

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

medium Chat template differs from claimed parent

The chat template does not match Qwen/Qwen2.5-7B-Instruct'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.

  1. Run `ingot patch <owner/model>` — the patch manifest carries the parent's template and applies it to a local copy's `tokenizer_config.json`.
  2. 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.
  3. 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 7806 undertrained tokens (norm < 0.3× the vocabulary median of 0.794), including 185 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "PostalCodesNL", "<unk>", "(stypy", "$PostalCodesNL", "TokenNameIdentifier", "Cumhurba", "thuisontvangst", "ForCanBeConvertedToF". 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.

low Lineage vs claimed parent Qwen/Qwen2.5-7B-Instruct inconclusive

Mean embedding-row cosine similarity to the declared base is 0.590 — below the 0.8 typical of true derivatives but not low enough to call mislabeled. Heavy continued pretraining or vocabulary surgery can look like this; verify provenance before relying on the parent's safety or licensing posture.

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.

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 KeefeBuild/Keefe-Discere

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.

architectureqwen2 · 28 layers · 3584-dim
parameters7615.6M
vocabulary152,064 tokens
licenseapache-2.0
serializationsafetensors + gguf
chat templatepresent · sha256:93ece0037c37d080
claimed lineageQwen/Qwen2.5-7B-Instruct
lineage verifiedinconclusive vs Qwen/Qwen2.5-7B-Instruct — embedding-row cosine 0.590
glitch-token surface7,806 undertrained candidates, 185 plain-ASCII
Full measured fingerprint
architecturesQwen2ForCausalLM
librarytransformers
pipelinetext-generation
repo files17
revisioncd1aff297c29
HF snapshot1.4k downloads · 1 likes · updated 2026-08-18 · captured 2026-08-21
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
embedding normsmedian 0.7936 · mean 0.7313
lineage checkinconclusive — cosine 0.5897 over 64 sampled rows vs Qwen/Qwen2.5-7B-Instruct
glitch-token samples"PostalCodesNL", "<unk>", "(stypy", "$PostalCodesNL", "TokenNameIdentifier", "Cumhurba", "thuisontvangst", "ForCanBeConvertedToF", "prostituerte", "NdrFc", "-vesm", "aincontri"
Battery runs (1)the run trace behind the findings — what each job measured
batterystatusqueueddurationattempts
weightscomplete2026-08-21 05:1283s1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
glitch surface7,806 undertrained, 185 plain-ASCII
lineage checkinconclusive — cosine 0.5897 over 64 rows vs Qwen/Qwen2.5-7B-Instruct

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/KeefeBuild/Keefe-Discere/badge.svg)](https://ingot.tools/models/KeefeBuild/Keefe-Discere)
Gate it in CI