purewhite42/rautoformalizer_ra_internlm warn
Weights only ship in a format that can run code when loaded; loading it runs custom code from the repo; its license differs from its base model's. Plus 1 more issue.
claims base: internlm/internlm2-math-base-7b, deepseek-ai/deepseek-math-7b-base · chat template: present · view on Hugging Face ↗
Scan coverageStatic battery2026-08-26Weights battery2026-08-26Behavioral batterynot rundetails
| Battery | Looks at | Status |
|---|---|---|
| Static battery | Metadata & packaging | complete 2026-08-26 |
| Weights battery | Weights forensics: no GPU, no download | complete 2026-08-2692,544-token embedding scanned · 642 undertrained · lineage consistent · pickle audit clean |
| Behavioral battery | Live-inference differentials | not run |
Ingot runs three batteries against a model. What each one checks →
Findings
Scanned 2026-08-26 · published from a community scan.
medium Pickle-serialized weights, no safetensors
Weights ship only as pickle-based files (pytorch_model-00001-of-00008.bin, pytorch_model-00002-of-00008.bin, pytorch_model-00003-of-00008.bin, …). Loading pickle executes arbitrary code from the file — prefer a safetensors release or load in a sandbox.
How to fix
Convert the weights to safetensors before loading them anywhere that matters.
- Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
- Convert locally in a sandbox: `pip install safetensors` and use `safetensors.torch.save_file` on a state dict loaded with `torch.load(..., weights_only=True)` (refuses most code-execution payloads), or use Hugging Face's `convert.py` space/script.
- Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.
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.
low Padding token is the EOS token
The pad token and the (only) EOS token are the same. Fine-tuning frameworks mask pad positions out of the loss, so training on this checkpoint teaches the model to never emit EOS — the Phi-4 / Qwen 2.5 / DeepSeek R1 infinite-generation bug. Safe to serve, hazardous to fine-tune; repoint pad_token at a dedicated token first.
How to fixingot patch
Align the stop-token declarations — a pure metadata fix to `generation_config.json` (and `config.json`).
- Identify the token the chat template actually ends assistant turns with (e.g. `<|eot_id|>`, `<end_of_turn>`, `<|im_end|>`) and make sure its id is in `generation_config.json`'s `eos_token_id` list.
- Keep `config.json`'s `eos_token_id` consistent with (or a subset of) `generation_config.json`'s — runtimes differ in which file they read.
- For the pad-equals-EOS hazard: repoint `pad_token` at a dedicated padding token before fine-tuning; serving is unaffected.
- Until the repo is fixed, pass explicit stop tokens to your serving stack (e.g. vLLM `stop_token_ids`, llama.cpp `--override-kv tokenizer.ggml.eos_token_id`).
medium License differs from claimed parent (apache-2.0 vs other)
This model declares apache-2.0 while its claimed base internlm/internlm2-math-base-7b declares other. 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.
medium Undertrained (glitch) token surface in vocabulary
Embedding-norm scan flagged 642 undertrained tokens (norm < 0.3× the vocabulary median of 0.996), including 266 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "[UNUSED_TOKEN_104]", "[UNUSED_TOKEN_140]", "[UNUSED_TOKEN_102]", "[UNUSED_TOKEN_120]", "[UNUSED_TOKEN_124]", "[UNUSED_TOKEN_129]", "[UNUSED_TOKEN_125]", "[UNUSED_TOKEN_128]". 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.
- 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.
- 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.
- 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.
info Pickle static analysis clean
Opcode-level parse of pytorch_model-00001-of-00008.bin, pytorch_model-00002-of-00008.bin, pytorch_model-00003-of-00008.bin, pytorch_model-00004-of-00008.bin (no code executed) found only standard serialization globals (3 distinct, all torch/collections/numpy). Pickle remains an executable format — this verifies the current bytes, not future uploads; prefer a safetensors release.
info Weights consistent with claimed parent internlm/internlm2-math-base-7b
Mean cosine similarity of 64 sampled token-embedding rows against internlm/internlm2-math-base-7b is 1.000 — the weights plausibly descend from the declared base (relation: unspecified).
Check every checkpoint before it ships
Use the web app, API, CLI, or CI gate to scan candidate checkpoints and catch model drift before deployment. Public-model scans publish to the open database; paid plans add the volume needed for continuous checks.
Fix it
Some findings are metadata-level and patchable — apply the fixes to your local copy (your weights never leave your machine):
npx @ingotai/scan patch purewhite42/rautoformalizer_ra_internlm
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 2026-08-26.
| architecture | internlm2 · 32 layers · 4096-dim |
| vocabulary | 92,544 tokens |
| license | apache-2.0 |
| serialization | no safetensors pickle custom code |
| chat template | present · sha256:49aa7cb516942cc5 |
| claimed lineage | internlm/internlm2-math-base-7b, deepseek-ai/deepseek-math-7b-base |
| lineage verified | consistent vs internlm/internlm2-math-base-7b — embedding-row cosine 1.000 |
| glitch-token surface | 642 undertrained candidates, 266 plain-ASCII |
Full measured fingerprint
| architectures | InternLM2ForCausalLM |
| library | transformers |
| pipeline | text-generation |
| repo files | 21 — pickle: pytorch_model-00001-of-00008.bin, pytorch_model-00002-of-00008.bin, pytorch_model-00003-of-00008.bin, pytorch_model-00004-of-00008.bin, pytorch_model-00005-of-00008.bin, pytorch_model-00006-of-00008.bin, pytorch_model-00007-of-00008.bin, pytorch_model-00008-of-00008.bin |
| revision | b6fb014a2814 |
| HF snapshot | 14 downloads · 0 likes · updated 2025-05-20 · captured 2026-08-25 |
| pickle audit | pytorch_model-00001-of-00008.bin, pytorch_model-00002-of-00008.bin, pytorch_model-00003-of-00008.bin, pytorch_model-00004-of-00008.bin — 3 standard global(s) |
| embedding tensor | model.tok_embeddings.weight · F16 · 92,544×4096 |
| embedding norms | median 0.9962 · mean 0.9563 |
| lineage check | consistent — cosine 1 over 64 sampled rows vs internlm/internlm2-math-base-7b |
| glitch-token samples | "[UNUSED_TOKEN_104]", "[UNUSED_TOKEN_140]", "[UNUSED_TOKEN_102]", "[UNUSED_TOKEN_120]", "[UNUSED_TOKEN_124]", "[UNUSED_TOKEN_129]", "[UNUSED_TOKEN_125]", "[UNUSED_TOKEN_128]", "[UNUSED_TOKEN_108]", "[UNUSED_TOKEN_103]", "[UNUSED_TOKEN_130]", "[UNUSED_TOKEN_114]" |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-25 22:32 | 2m | 1 |
weights run 2026-08-25 — measurements
| probes run | glitch-norm-scan, zero-template-token-scan, pickle-static-analysis, lineage-norm-correlation |
| embedding tensor | model.tok_embeddings.weight · F16 · 92,544×4096 |
| glitch surface | 642 undertrained, 266 plain-ASCII |
| lineage check | consistent — cosine 1 over 64 rows vs internlm/internlm2-math-base-7b |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/purewhite42/rautoformalizer_ra_internlm)