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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Friends
    -0.07
     perhaps
    -0.07
     aktiv
    -0.06
    яем
    -0.06
    '),
    -0.06
    ,H
    -0.06
    ambia
    -0.06
     serve
    -0.06
    Meanwhile
    -0.06
    泉
    -0.06
    POSITIVE LOGITS
    LTE
    0.07
    গ
    0.07
    лон
    0.07
     admissions
    0.07
     handguns
    0.07
    .IDENTITY
    0.06
     Kore
    0.06
     sürekli
    0.06
    溵
    0.06
    iete
    0.06
    Act Density 0.002%

    No Known Activations