keyfan/vicuna-chinese-replication-v1.1 warn
Weights only ship in a format that can run code when loaded; EOS ids disjoint between config.json and generation_config.json; Special token id outside the vocabulary. Plus 1 minor note.
Could not load this model from the Hugging Face API (private, gated, or nonexistent). Findings below are from our archive.
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 2026-08-2649,953-token embedding scanned · 172 undertrained · 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-00003.bin, pytorch_model-00002-of-00003.bin, pytorch_model-00003-of-00003.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 EOS ids disjoint between config.json and generation_config.json
config.json declares eos_token_id [2] while generation_config.json declares [1] with no overlap. Runtimes read one or the other, so at least one of them stops generation on the wrong token (or never). Align both files on the token the chat template actually ends turns with.
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 Special token id outside the vocabulary
pad_token_id=-1 is outside the declared vocab_size of 49953. Runtimes either crash on it or silently ignore the setting (e.g. pad_token_id: -1), which breaks batching and stop handling.
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`).
low Undertrained tokens in vocabulary (non-ASCII tail)
Embedding-norm scan flagged 172 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.
- 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-00003.bin, pytorch_model-00002-of-00003.bin, pytorch_model-00003-of-00003.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.
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 keyfan/vicuna-chinese-replication-v1.1
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 | llama · 40 layers · 5120-dim |
| vocabulary | 49,953 tokens |
| license | other |
| serialization | no safetensors pickle |
| chat template | none |
| glitch-token surface | 172 undertrained candidates, 0 plain-ASCII |
Full measured fingerprint
| architectures | LlamaForCausalLM |
| library | transformers |
| pipeline | text-generation |
| repo files | 10 — pickle: pytorch_model-00001-of-00003.bin, pytorch_model-00002-of-00003.bin, pytorch_model-00003-of-00003.bin |
| revision | 259ab0967975 |
| HF snapshot | 114 downloads · 4 likes · updated 2023-05-23 · captured 2026-08-25 |
| pickle audit | pytorch_model-00001-of-00003.bin, pytorch_model-00002-of-00003.bin, pytorch_model-00003-of-00003.bin — 3 standard global(s) |
| embedding tensor | model.embed_tokens.weight · F16 · 49,953×5120 |
| embedding norms | median 2.2164 · mean 2.09 |
| lineage check | no claimed base model |
Battery runs (1)the run trace behind the findings — what each job measured
| battery | status | queued | duration | attempts |
|---|---|---|---|---|
| weights | complete | 2026-08-25 21:52 | 2m | 1 |
weights run 2026-08-25 — measurements
| probes run | glitch-norm-scan, zero-template-token-scan, pickle-static-analysis, lineage-norm-correlation |
| probes skipped | token-decode: no tokenizer.json |
| embedding tensor | model.embed_tokens.weight · F16 · 49,953×5120 |
| glitch surface | 172 undertrained, 0 plain-ASCII |
| lineage check | not checked (no claimed base model) |
Verdict badge
Ship the verdict in your README — it always shows the latest published analysis:
[](https://ingot.tools/models/keyfan/vicuna-chinese-replication-v1.1)