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

    Neuronpedia

    INDEX
    Explanations

    maybe? smooth. sometimes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Dieses
    0.47
     Dieses
    0.45
     This
    0.45
    Diese
    0.44
     continuamente
    0.44
    Ist
    0.43
    This
    0.43
    /
    0.42
     These
    0.42
     (
    0.41
    POSITIVE LOGITS
     fáciles
    0.54
     الهمزه
    0.51
    ですね
    0.50
     женой
    0.50
    이죠
    0.49
     минус
    0.48
     बढ़िया
    0.48
     없고
    0.47
    군요
    0.47
     숫
    0.47
    Act Density 0.033%

    No Known Activations