© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 25268
    Prev
    Next
    INDEX
    Explanations

    CVEs and security vulnerabilities

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires strongly on mentions of software security flaws and their identifiers (e.g. “vulnerability,” “CVE,” zero-day, bug, patch), thus detecting references to security vulnerabilities.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ׇ
    -1.63
     my
    -1.49
     گیاه
    -1.30
    apatalk
    -1.25
     Pfar
    -1.20
    ֺ
    -1.19
     растение
    -1.18
     I
    -1.18
     ингреди
    -1.18
     buena
    -1.13
    POSITIVE LOGITS
    וּ
    2.03
    ᾿
    1.56
    ּוֹ
    1.48
    וֹ
    1.43
    ᾽
    1.42
     شکر
    1.38
    ַּ
    1.35
     freundlichen
    1.34
    וּ
    1.34
    ְּ
    1.33
    Activations Density 0.043%

    No Known Activations