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

    Neuronpedia

    INDEX
    Explanations

    potential problems/negative consequences

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    enate
    -0.07
     unavoid
    -0.07
     모집
    -0.06
     Nacht
    -0.06
    ey
    -0.06
    observe
    -0.06
    Sharper
    -0.06
    旁
    -0.06
     사이
    -0.06
    shit
    -0.06
    POSITIVE LOGITS
     enable
    0.07
     onze
    0.06
     %=
    0.06
    ả
    0.06
    ,class
    0.06
    нова
    0.06
    ा।
    0.06
    mızı
    0.06
    ANTI
    0.06
    (signature
    0.06
    Act Density 0.211%

    No Known Activations