Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    constructor definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    CSE
    0.44
    システムの
    0.38
    市民
    0.38
     systems
    0.38
     სისტ
    0.38
    వి
    0.36
     SYSTEM
    0.36
    龊
    0.36
    য়ার
    0.36
     civ
    0.35
    POSITIVE LOGITS
    ToPointer
    0.40
    ophage
    0.39
     팔
    0.39
     sinc
    0.39
     morphine
    0.37
     ModInt
    0.37
     salvage
    0.37
     outweighs
    0.37
    resser
    0.36
     Anschluss
    0.36
    Act Density 0.064%

    No Known Activations