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

    Neuronpedia

    INDEX
    Explanations

    I always do

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    果た
    0.77
     Таким
    0.71
     तुमचे
    0.70
     Aeronaut
    0.70
     तुमच्या
    0.69
     phát
    0.69
     فرح
    0.69
    داث
    0.68
    发射
    0.68
     perpetrated
    0.68
    POSITIVE LOGITS
     always
    1.15
    always
    1.15
     siempre
    1.09
    注重
    1.03
    我会
    1.01
     altid
    0.99
     alltid
    0.97
    Always
    0.97
     Always
    0.96
    ALWAYS
    0.95
    Act Density 0.039%

    No Known Activations