© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 49618
    Prev
    Next
    INDEX
    Explanations

    religious and professional titles followed by names

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

    The neuron consistently activates on formal titles or honorifics (e.g., “Rabbi,” “Shaykh,” “Imam,” “Father,” “Judge”) preceding names.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     kabát
    -1.29
    
    -1.29
     TAMBÉM
    -1.28
     kré
    -1.27
     pól
    -1.23
     oferty
    -1.20
     luxus
    -1.19
     drugi
    -1.16
    蹀
    -1.15
     komfor
    -1.13
    POSITIVE LOGITS
     Mike
    1.30
     годом
    1.23
     矢量
    1.23
     Robert
    1.23
    際には
    1.21
     there
    1.20
     Emeritus
    1.18
     Michael
    1.18
     Peter
    1.17
     Andrew
    1.16
    Activations Density 0.052%

    No Known Activations