Model page

danish-foundation-models/DFM-Mimir warn

Glitch tokens that can silently corrupt ordinary input.

downloads 5.1klikes 51license apache-2.0arch hrm_textparams 1786.8Mupdated 2026-08-20

chat template: present · view on Hugging Face ↗

Scan coverageStatic battery2026-08-25Weights battery2026-08-25Behavioral batterynot rundetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-25
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-25262,144-token embedding scanned · 38050 undertrained
Behavioral batteryLive-inference differentialsnot run

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

Findings

Scanned 2026-08-25 · published from a community scan.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 38050 undertrained tokens (norm < 0.3× the vocabulary median of 0.705), including 11917 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "<unused4598>", "<unused147>", "btnbrowser", "<unused873>", "<unused2489>", "HistoryMarks", "<unused1874>", "<unused130>". 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.

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 2026-08-25.

architecturehrm_text · 16 layers · 1536-dim
parameters1786.8M
vocabulary262,144 tokens
licenseapache-2.0
serializationsafetensors
chat templatepresent (chat_template.jinja) · sha256:33204f1acb5bd000
glitch-token surface38,050 undertrained candidates, 11,917 plain-ASCII
Full measured fingerprint
architecturesHrmTextForCausalLM
librarytransformers
pipelinetext-generation
repo files10
revision2844f0178e69
HF snapshot5.1k downloads · 50 likes · updated 2026-08-20 · captured 2026-08-25
embedding tensormodel.embed_tokens.weight · BF16 · 262,144×1536
embedding normsmedian 0.7048 · mean 0.6899
lineage checkno claimed base model
glitch-token samples"<unused4598>", "<unused147>", "btnbrowser", "<unused873>", "<unused2489>", "HistoryMarks", "<unused1874>", "<unused130>", "<unused289>", "<unused5881>", "<unused3110>", "<unused683>"
Battery runs (1)the run trace behind the findings — what each job measured
batterystatusqueueddurationattempts
weightscomplete2026-08-25 19:0033s1
weights run 2026-08-25 measurements
probes runglitch-norm-scan, zero-template-token-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 262,144×1536
glitch surface38,050 undertrained, 11,917 plain-ASCII
lineage checknot checked (no claimed base model)

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/danish-foundation-models/DFM-Mimir/badge.svg)](https://ingot.tools/models/danish-foundation-models/DFM-Mimir)
Gate it in CI