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

    Neuronpedia

    INDEX
    Explanations

    explaining reasons why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     जायेगा
    0.49
     должна
    0.42
     deberá
    0.40
    一定会
    0.40
    Trashed
    0.39
    也會
    0.39
     powinien
    0.39
     potrà
    0.38
     cps
    0.38
     deverá
    0.38
    POSITIVE LOGITS
     omdat
    0.68
     porque
    0.66
    porque
    0.64
     क्योंकि
    0.62
    because
    0.62
     Çünkü
    0.60
     einfach
    0.59
     BECAUSE
    0.59
     때문이다
    0.59
     Reasons
    0.59
    Act Density 0.219%

    No Known Activations