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

    Neuronpedia

    INDEX
    Explanations

    det

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    phalodes
    0.36
    やま
    0.34
    !='')
    0.34
     eld
    0.33
    бак
    0.33
     hiyo
    0.32
     柄
    0.32
    ácil
    0.31
     hina
    0.31
     મળે
    0.31
    POSITIVE LOGITS
     det
    4.09
    Det
    3.41
     Det
    3.28
    det
    3.23
     dét
    2.77
     DET
    2.56
    DET
    2.42
     Dét
    2.17
     dete
    1.87
    dets
    1.85
    Act Density 0.014%

    No Known Activations