Model page

Qwen/Qwen2-7B warn

Chat template ends turns with <|im_end|>, which is not a configured stop token; glitch tokens that can silently corrupt ordinary input. 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 battery2026-08-25Weights battery2026-08-25Behavioral batteryfaileddetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-25
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-25152,064-token embedding scanned · 7405 undertrained
Behavioral batteryLive-inference differentialsfailedTraceback (most recent call last): | raise HTTPStatusError(message, request=request, response=self) | httpx.HTTPStatusError: Client error '429 Too Many Requests' for url 'https://huggingface.co/api/models/Qwen/Qwen2-7B' | Traceback (most recent call last): | raise _format(HfHubHTTPError, message, response) from e | huggingface_hub.errors.HfHubHTTPError: (Request ID: Root=1-6a8e107c-0193ed890ad33f6e11b88f09;53ab449c-2552-46bf-b0d9-d143c3aaf802)

Ingot runs three batteries against a model. What each one checks →

Findings

Scanned 2026-08-25 · published from a community scan.

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`).

  1. 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.
  2. Keep `config.json`'s `eos_token_id` consistent with (or a subset of) `generation_config.json`'s — runtimes differ in which file they read.
  3. For the pad-equals-EOS hazard: repoint `pad_token` at a dedicated padding token before fine-tuning; serving is unaffected.
  4. 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 Chat template ends turns with <|im_end|>, which is not a configured stop token

The chat template terminates assistant turns with <|im_end|>, but the effective EOS set (config.json ∪ generation_config.json = [151643] → ["<|endoftext|>"]) never stops on it. Config-honoring runtimes generate past the terminator until the token budget is exhausted — runaway cost and self-continuing fake turns. Add <|im_end|>'s id to generation_config.json's eos_token_id.

How to fixingot patch

Align the stop-token declarations — a pure metadata fix to `generation_config.json` (and `config.json`).

  1. 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.
  2. Keep `config.json`'s `eos_token_id` consistent with (or a subset of) `generation_config.json`'s — runtimes differ in which file they read.
  3. For the pad-equals-EOS hazard: repoint `pad_token` at a dedicated padding token before fine-tuning; serving is unaffected.
  4. 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 Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 7405 undertrained tokens (norm < 0.3× the vocabulary median of 0.742), including 215 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "ForCanBeConverted", "ForCanBeConvertedToF", "PostalCodesNL", "$PostalCodesNL", "useRalative", "<unk>", "(stypy", "useRal". 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.

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 Qwen/Qwen2-7B

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 2026-08-25.

architectureqwen2 · 28 layers · 3584-dim
parameters7615.6M
vocabulary152,064 tokens
licenseapache-2.0
serializationsafetensors
chat templatepresent · sha256:af9c0233881b083b
glitch-token surface7,405 undertrained candidates, 215 plain-ASCII
Full measured fingerprint
architecturesQwen2ForCausalLM
librarytransformers
pipelinetext-generation
repo files14
revision453ed1575b73
HF snapshot222.5k downloads · 174 likes · updated 2024-06-06 · captured 2026-08-25
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
embedding normsmedian 0.7419 · mean 0.6759
lineage checkno claimed base model
glitch-token samples"ForCanBeConverted", "ForCanBeConvertedToF", "PostalCodesNL", "$PostalCodesNL", "useRalative", "<unk>", "(stypy", "useRal", "TokenNameIdentifier", "-vesm", "methodPointerType", "webElementX"
Battery runs (1)the run trace behind the findings — what each job measured
batterystatusqueueddurationattempts
weightscomplete2026-08-25 20:3446s1
weights run 2026-08-25 measurements
probes runglitch-norm-scan, zero-template-token-scan, lineage-norm-correlation
embedding tensormodel.embed_tokens.weight · BF16 · 152,064×3584
glitch surface7,405 undertrained, 215 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/Qwen/Qwen2-7B/badge.svg)](https://ingot.tools/models/Qwen/Qwen2-7B)
Gate it in CI