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

    Neuronpedia

    INDEX
    Explanations

    say "const"

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    差
    0.67
     ری
    0.67
    絡
    0.63
     clubs
    0.62
    전을
    0.61
    都
    0.61
     hub
    0.61
     HUB
    0.61
     மாற்றம்
    0.60
     বিশ্র
    0.60
    POSITIVE LOGITS
    const
    0.78
     jego
    0.73
    expandedTitle
    0.70
     //=>
    0.68
    റിന്റെ
    0.68
     रिंग
    0.68
     texId
    0.67
     njegove
    0.66
    ಕೊ
    0.65
     จาก
    0.65
    Act Density 0.006%

    No Known Activations