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

    Neuronpedia

    INDEX
    Explanations

    words before or after specific tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    invest
    0.48
     Продол
    0.44
     Parte
    0.43
    Arte
    0.43
    arine
    0.42
     awal
    0.41
     наи
    0.41
    াইয়
    0.41
     آرام
    0.41
     சிறிய
    0.41
    POSITIVE LOGITS
    elementType
    0.45
    ’।
    0.45
     Donnelly
    0.44
    ্নে
    0.44
     grandson
    0.42
    تقديم
    0.41
    ⓡ
    0.40
    ਗੇ
    0.40
    네
    0.39
    숲
    0.39
    Act Density 0.005%

    No Known Activations