Model page

google/pegasus-xsum warn

Weights only ship in a format that can run code when loaded; no license declared — no usage rights by default; glitch tokens that can silently corrupt ordinary input.

downloads 168.8klikes 222license none declaredarch pegasusupdated 2023-01-24

chat template: not found · view on Hugging Face ↗

Scan coverageStatic battery2026-08-25Weights battery2026-08-25Behavioral batterynot rundetails
BatteryLooks atStatus
Static batteryMetadata & packagingcomplete 2026-08-25
Weights batteryWeights forensics — no GPU, no downloadcomplete 2026-08-2596,103-token embedding scanned · 7543 undertrained · pickle audit clean
Behavioral batteryLive-inference differentialsnot run

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

Findings

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

medium Pickle-serialized weights, no safetensors

Weights ship only as pickle-based files (pytorch_model.bin, tf_weights_dict.pkl). 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.

  1. Do not load the pickle files in-process — pickle deserialization executes arbitrary code from the file.
  2. 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.
  3. Pin the exact revision hash you converted from, and load only your converted safetensors artifact from then on.

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 7543 undertrained tokens (norm < 0.3× the vocabulary median of 14.547), including 6116 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "HALF", "apotheke", "Obtain", "ANTI", "parsed", "Gazebo", "labeled", "MARK". 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.

info Pickle static analysis clean

Opcode-level parse of pytorch_model.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.

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.

architecturepegasus · 16 layers · 1024-dim
vocabulary96,103 tokens
licensenone declared
serializationno safetensors pickle
chat templatenone
glitch-token surface7,543 undertrained candidates, 6,116 plain-ASCII
Full measured fingerprint
architecturesPegasusForConditionalGeneration
librarytransformers
pipelinesummarization
repo files12 — pickle: pytorch_model.bin, tf_weights_dict.pkl
revision8d8ffc158a3b
HF snapshot168.8k downloads · 222 likes · updated 2023-01-24 · captured 2026-08-25
pickle auditpytorch_model.bin3 standard global(s) · legacy (pre-1.6) format, head-scan only
embedding tensormodel.shared.weight · F32 · 96,103×1024
embedding normsmedian 14.5467 · mean 13.8004
lineage checkno claimed base model
glitch-token samples"HALF", "apotheke", "Obtain", "ANTI", "parsed", "Gazebo", "labeled", "MARK", "Boda", "Angora", "CONCEPT", "POINT"
Battery runs (1)the run trace behind the findings — what each job measured
batterystatusqueueddurationattempts
weightscomplete2026-08-25 20:2619s1
weights run 2026-08-25 measurements
probes runglitch-norm-scan, zero-template-token-scan, pickle-static-analysis, lineage-norm-correlation
embedding tensormodel.shared.weight · F32 · 96,103×1024
glitch surface7,543 undertrained, 6,116 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/google/pegasus-xsum/badge.svg)](https://ingot.tools/models/google/pegasus-xsum)
Gate it in CI