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

    Neuronpedia

    INDEX
    Explanations

    json format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    َرَ
    0.82
    Ét
    0.80
    related
    0.79
    ңуз
    0.79
     बहुतेक
    0.75
    /
    0.75
    themed
    0.74
    !),
    0.71
    ங்களைப்
    0.70
    études
    0.69
    POSITIVE LOGITS
     atleast
    1.97
     upto
    1.89
     irrespective
    1.57
     loosing
    1.57
     Hence
    1.48
     squre
    1.47
     :-
    1.47
     까지
    1.45
     inorder
    1.43
     입니다
    1.42
    Act Density 0.097%

    No Known Activations