Model page

deepseek-ai/DeepSeek-R1-0528-Qwen3-8B warn

downloads 1.2Mlikes 1.1klicense mitarch qwen3params 8190.7Mupdated 2025-05-29

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 2026-08-22
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-22151,936-token embedding scanned · 3011 undertrained
Behavioral batteryLive-inference differentialsnot run

Findings

Scanned 2026-08-22 · published from a community scan.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 3011 undertrained tokens (norm < 0.3× the vocabulary median of 1.472), including 104 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "$PostalCodesNL", "ForCanBeConvertedToF", "ForCanBeConverted", "PostalCodesNL", "useRalative", "useRal", "thuisontvangst", "sexkontakte". 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 2026-08-22.

architectureqwen3 · 36 layers · 4096-dim
parameters8190.7M
vocabulary151,936 tokens
licensemit
serializationsafetensors
chat templatepresent · sha256:53671bac29bcb3e1
glitch-token surface3,011 undertrained candidates, 104 plain-ASCII
Full measured fingerprint
architecturesQwen3ForCausalLM
librarytransformers
pipelinetext-generation
repo files10
revision6e8885a6ff5c
HF snapshot1.2M downloads · 1.1k likes · updated 2025-05-29 · captured 2026-08-21
embedding tensormodel.embed_tokens.weight · BF16 · 151,936×4096
embedding normsmedian 1.4724 · mean 1.3963
lineage checkno claimed base model
glitch-token samples"$PostalCodesNL", "ForCanBeConvertedToF", "ForCanBeConverted", "PostalCodesNL", "useRalative", "useRal", "thuisontvangst", "sexkontakte", "NdrFc", "webElementX", "sextreffen", "wannonce"

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 07:4362s1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 151,936×4096
glitch surface3,011 undertrained, 104 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/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B/badge.svg)](https://ingot.tools/models/deepseek-ai/DeepSeek-R1-0528-Qwen3-8B)
Gate it in CI