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

    Neuronpedia

    INDEX
    Explanations

    door and door-related terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    리
    1.38
    ۔
    1.25
    ين
    1.18
     speichern
    1.16
    은
    1.16
    트
    1.09
    ח
    1.08
    י
    1.05
    は
    1.05
     dargestellt
    1.03
    POSITIVE LOGITS
    ки
    1.09
     Door
    1.05
     and
    1.02
    on
    1.01
    وم
    0.97
    S
    0.96
    <0x0D>
    0.95
    at
    0.95
    ana
    0.95
    aj
    0.94
    Act Density 0.005%

    No Known Activations