Model page

trl-internal-testing/tiny-Qwen2ForCausalLM-2.5 warn

downloads 16.4Mlikes 20license none declaredarch qwen2params 2.4Mupdated 2025-12-19

chat template: not in config · view on Hugging Face ↗

Ingot findings

Static battery: 1 medium finding(s). Deep battery (behavioral differential, glitch-token pass) not yet run. Weights battery: embedding-norm scan over 151665 tokens (BF16, 8-dim) found 48 undertrained candidates, 19 plain-ASCII. Scanned 2026-08-20 (published from a community scan).

medium No license declared

The model card declares no license. You have no usage rights by default — treat as all-rights-reserved until the owner clarifies.

How to fix

Get a license from the owner or pick a licensed alternative — this is a legal gap, not a technical one.

  1. With no declared license you have no usage rights by default; treat the weights as all-rights-reserved.
  2. Open an issue or discussion on the repo asking the owner to declare a license, or use the licensed upstream/parent model instead.

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 48 undertrained tokens (norm < 0.3× the vocabulary median of 0.054), including 19 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "stems", "tracing", "fleet", "aters", ".pojo", "izard", "-centric", "integrates". 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 GPU deep 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.

Remediation guidance addresses the documented findings only. It is evidence-driven repair, not a safety certification of the model.

Fingerprint

The durable weights-and-metadata profile of this model, rebuilt on every scan and deep-battery run. Updated 2026-08-20.

architectureqwen2 · 2 layers · 8-dim
parameters2.4M
vocabulary151,665 tokens
licensenone declared
serializationsafetensors
chat templatenone
glitch-token surface48 undertrained candidates, 19 plain-ASCII
Full fingerprint
architecturesQwen2ForCausalLM
librarytransformers
pipelinetext-generation
repo files12
revision4b10ebee6e13
HF snapshot16.4M downloads · 20 likes · updated 2025-12-19 · captured 2026-08-20
embedding tensormodel.embed_tokens.weight · BF16 · 151,665×8
embedding normsmedian 0.054 · mean 0.0546
lineage checkno claimed base model
glitch-token samples"stems", "tracing", "fleet", "aters", ".pojo", "izard", "-centric", "integrates", "stdout", "clud", "circle", "promo"

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/trl-internal-testing/tiny-Qwen2ForCausalLM-2.5/badge.svg)](https://ingot.tools/models/trl-internal-testing/tiny-Qwen2ForCausalLM-2.5)
Gate it in CI