MERaLiON/MERaLiON-2-10B warn
claims base: openai/whisper-large-v3, google/gemma-2-9b-it · chat template: present · view on Hugging Face ↗
Scan coverage
Ingot runs three batteries against a model. What each one checks →
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-21 |
| Weights battery | Weights forensics — no GPU, no download | complete 2026-08-21256,000-token embedding scanned · 0 undertrained · lineage inconsistent |
| Behavioral battery | Live-inference differentials | n/anot applicable — automatic-speech-recognition model has no text-generation surface to probe |
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.
- Read every `.py` file in the repo before first load — this code runs in your process.
- 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.
- 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 (other vs apache-2.0)
This model declares other while its claimed base openai/whisper-large-v3 declares apache-2.0. Verify the re-license is permitted before commercial use.
How to fix
Verify the re-license is actually permitted before relying on it.
- 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.
- If the re-license is not permitted, the parent's terms govern your use regardless of what this repo declares.
info Embedding-norm glitch scan clean
No undertrained tokens found: every non-special token's embedding norm is above 0.3× the vocabulary median (1.696). The glitch-token data-corruption class has no candidate surface in this model.
medium Weights inconsistent with claimed parent openai/whisper-large-v3
This model declares openai/whisper-large-v3 as its base (relation: unspecified), but its token-embedding geometry is incompatible: 3584-dim embeddings vs the parent's 1280-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.
- If you own the repo: correct the `base_model` field in the model card metadata to the real, public parent.
- 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.
| architecture | meralion2 · 42 layers · 3584-dim |
| parameters | 10106.4M |
| vocabulary | 256,000 tokens |
| license | other |
| serialization | safetensors custom code |
| chat template | present · sha256:cd00f098978cd500 |
| claimed lineage | openai/whisper-large-v3, google/gemma-2-9b-it |
| lineage verified | inconsistent vs openai/whisper-large-v3 |
| glitch-token surface | clean no undertrained tokens |
Full measured fingerprint
| architectures | MERaLiON2ForConditionalGeneration |
| library | transformers |
| pipeline | automatic-speech-recognition |
| repo files | 22 |
| revision | 57bb08fd110a |
| HF snapshot | 301 downloads · 12 likes · updated 2026-03-31 · captured 2026-08-21 |
| embedding tensor | text_decoder.model.embed_tokens.weight · BF16 · 256,000×3584 |
| embedding norms | median 1.6959 · mean 1.7164 |
| lineage check | inconsistent — cosine undefined over undefined sampled rows vs openai/whisper-large-v3 |
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.
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-21 05:15 | 2m | 1 |
weights run 2026-08-21 — measurements
| probes run | glitch-norm-scan, lineage-norm-correlation |
| embedding tensor | text_decoder.model.embed_tokens.weight · BF16 · 256,000×3584 |
| glitch surface | 0 undertrained, 0 plain-ASCII |
| lineage check | inconsistent — cosine undefined over undefined rows vs openai/whisper-large-v3 |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/MERaLiON/MERaLiON-2-10B)