Model page

FacebookAI/roberta-large warn

downloads 11.1Mlikes 319license mitarch robertaparams 355.4Mupdated 2024-02-19

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

Ingot findings

Static battery clean: safetensors weights, license declared, no template/tokenizer drift detected. Deep battery not yet run. Weights battery: embedding-norm scan over 50265 tokens (F32, 1024-dim) found 175 undertrained candidates, 97 plain-ASCII. Scanned 2026-08-20 (published from a community scan).

medium Undertrained (glitch) token surface in vocabulary

Embedding-norm scan flagged 175 undertrained tokens (norm < 0.3× the vocabulary median of 4.489), including 97 plain-ASCII strings that can appear in ordinary input as identifiers — e.g. "PsyNetMessage", "FactoryReloaded", "TheNitrome", "Adinida", "channelAvailability", "NetMessage", "attRot", "Orderable". 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.

architectureroberta · 24 layers · 1024-dim
parameters355.4M
vocabulary50,265 tokens
licensemit
serializationsafetensors pickle
chat templatenone
glitch-token surface175 undertrained candidates, 97 plain-ASCII
Full fingerprint
architecturesRobertaForMaskedLM
librarytransformers
pipelinefill-mask
repo files12 — pickle: pytorch_model.bin
revision722cf37b1afa
HF snapshot11.1M downloads · 319 likes · updated 2024-02-19 · captured 2026-08-20
embedding tensorroberta.embeddings.word_embeddings.weight · F32 · 50,265×1024
embedding normsmedian 4.4894 · mean 4.3585
lineage checkno claimed base model
glitch-token samples"PsyNetMessage", "FactoryReloaded", "TheNitrome", "Adinida", "channelAvailability", "NetMessage", "attRot", "Orderable", "guiIcon", "isSpecialOrderable", "srfN", "guiActiveUn"

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/FacebookAI/roberta-large/badge.svg)](https://ingot.tools/models/FacebookAI/roberta-large)
Gate it in CI