Model page

deepseek-ai/deepseek-coder-7b-instruct-v1.5 warn

downloads 802.7klikes 160license otherarch llamaparams 6910.4Mupdated 2024-02-05

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 2026-08-22102,400-token embedding scanned · 5525 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 5525 undertrained tokens (norm < 0.3× the vocabulary median of 11.289), including 1729 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "IconSuccessEncoded", "IconErrorEncoded", "orangehilldev", "EDIPU", "lemanya", "odeciclismo", "RecordedVote", "textquoted". 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.

architecturellama · 30 layers · 4096-dim
parameters6910.4M
vocabulary102,400 tokens
licenseother
serializationsafetensors
chat templatepresent · sha256:993c0af704bad76a
glitch-token surface5,525 undertrained candidates, 1,729 plain-ASCII
Full measured fingerprint
architecturesLlamaForCausalLM
librarytransformers
pipelinetext-generation
repo files11
revision2a050a4c59d6
HF snapshot802.7k downloads · 160 likes · updated 2024-02-05 · captured 2026-08-21
embedding tensormodel.embed_tokens.weight · BF16 · 102,400×4096
embedding normsmedian 11.2889 · mean 10.4648
lineage checkno claimed base model
glitch-token samples"IconSuccessEncoded", "IconErrorEncoded", "orangehilldev", "EDIPU", "lemanya", "odeciclismo", "RecordedVote", "textquoted", "sympad", "linkedExternalProjectPath", "espany", "controlcap"

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 07:432m1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 102,400×4096
glitch surface5,525 undertrained, 1,729 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-coder-7b-instruct-v1.5/badge.svg)](https://ingot.tools/models/deepseek-ai/deepseek-coder-7b-instruct-v1.5)
Gate it in CI