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

    Neuronpedia

    INDEX
    Explanations

    servers and http

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Swain
    0.67
    isel
    0.65
    身
    0.61
     intervalles
    0.59
     woods
    0.58
    )}]{
    0.58
    ission
    0.57
     মুহ
    0.56
     insta
    0.56
    对待
    0.55
    POSITIVE LOGITS
    чески
    0.81
    <unused446>
    0.80
    gou
    0.78
     praktisch
    0.78
    㳖
    0.78
    klassen
    0.75
     Prakt
    0.74
     rund
    0.72
    рон
    0.71
     Rond
    0.71
    Act Density 0.001%

    No Known Activations