© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129455
    Prev
    Next
    INDEX
    Explanations

    The neuron detects pejorative, politically charged labels or compound insults (especially hyphenated terms) used to mock or caricature ideological opponents.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    biblical figures and entities

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     many
    -1.34
     often
    -1.16
     different
    -1.12
    ременно
    -1.04
     –
    -1.02
    </strong>
    -1.01
     sites
    -1.01
     similar
    -1.00
    ]};
    -0.99
     bot
    -0.99
    POSITIVE LOGITS
     السنة
    1.13
    sourceforge
    1.09
     herhangi
    1.07
    huur
    1.05
    setNome
    1.02
     akurat
    1.02
     fupp
    1.00
     NSCoder
    1.00
     geest
    0.99
    CTE
    0.99
    Activations Density 0.069%

    No Known Activations