Model page

nvidia/Eagle2-9B warn

downloads 330likes 63license cc-by-nc-4.0arch eagle_chatparams 8928.3Mupdated 2025-01-28

claims base: google/paligemma-3b-mix-448, Qwen/Qwen2.5-7B-Instruct, google/siglip-so400m-patch14-384, timm/convnext_xxlarge.clip_laion2b_soup_ft_in1k · 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 2026-08-21151,674-token embedding scanned · 6951 undertrained · lineage inconsistent
Behavioral batteryLive-inference differentialsnot run

Findings

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

medium Repo ships executable Python (trust_remote_code)

The repository contains custom code files that run in-process when loaded with trust_remote_code=True. Pin the revision hash and review the code before loading.

How to fix

Review and pin the custom code; never float on `main` with trust_remote_code=True.

  1. Read every `.py` file in the repo before first load — this code runs in your process.
  2. Pin the revision: `from_pretrained(model_id, revision="<commit sha>", trust_remote_code=True)` so a later push can't swap the code under you.
  3. Prefer a version of the architecture already in `transformers` if one exists, which removes the remote-code requirement entirely.

medium License differs from claimed parent (cc-by-nc-4.0 vs gemma)

This model declares cc-by-nc-4.0 while its claimed base google/paligemma-3b-mix-448 declares gemma. Verify the re-license is permitted before commercial use.

How to fix

Verify the re-license is actually permitted before relying on it.

  1. Read the parent's license for derivative-work and re-licensing terms — many open-weight licenses (e.g. Llama-family) do not permit arbitrary re-licensing.
  2. If the re-license is not permitted, the parent's terms govern your use regardless of what this repo declares.

low Undertrained tokens in vocabulary (non-ASCII tail)

Embedding-norm scan flagged 6951 undertrained tokens (norm < 0.3× the vocabulary median), but 0 decode to plain-ASCII strings, so exposure in English-language pipelines is limited. 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.

medium Weights inconsistent with claimed parent google/paligemma-3b-mix-448

This model declares google/paligemma-3b-mix-448 as its base (relation: unspecified), but its token-embedding geometry is incompatible: 3584-dim embeddings vs the parent's 2048-dim. A finetune cannot change embedding width — the lineage label is wrong or misleading. Treat provenance claims on this repo (training data, safety posture, licensing) as unverified.

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.

architectureeagle_chat
parameters8928.3M
vocabulary151,674 tokens
licensecc-by-nc-4.0
serializationsafetensors custom code
chat templatepresent · sha256:d5495a1e5db06111
claimed lineagegoogle/paligemma-3b-mix-448, Qwen/Qwen2.5-7B-Instruct, google/siglip-so400m-patch14-384, timm/convnext_xxlarge.clip_laion2b_soup_ft_in1k
lineage verifiedinconsistent vs google/paligemma-3b-mix-448
glitch-token surface6,951 undertrained candidates, 0 plain-ASCII
Full measured fingerprint
architecturesEagle2ChatModel
librarytransformers
pipelineimage-text-to-text
repo files31
revision3f112192f66f
HF snapshot333 downloads · 63 likes · updated 2025-01-28 · captured 2026-08-21
embedding tensorlanguage_model.model.embed_tokens.weight · BF16 · 151,674×3584
embedding normsmedian 0.8573 · mean 0.7911
lineage checkinconsistent — cosine undefined over undefined sampled rows vs google/paligemma-3b-mix-448

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 05:152m1
weights run 2026-08-21 measurements
probes runglitch-norm-scan, lineage-norm-correlation
probes skippedtoken-decode: no tokenizer.json
embedding tensorlanguage_model.model.embed_tokens.weight · BF16 · 151,674×3584
glitch surface6,951 undertrained, 0 plain-ASCII
lineage checkinconsistent — cosine undefined over undefined rows vs google/paligemma-3b-mix-448

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/nvidia/Eagle2-9B/badge.svg)](https://ingot.tools/models/nvidia/Eagle2-9B)
Gate it in CI