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

    Neuronpedia

    INDEX
    Explanations

    must or should

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    सेल
    0.69
    ุป
    0.66
    罷
    0.64
    дами
    0.61
     consigui
    0.61
    <unused1741>
    0.60
     keç
    0.59
    اویز
    0.58
     ይችላሉ
    0.58
    ไตล์
    0.58
    POSITIVE LOGITS
     should
    5.05
     Should
    4.47
    should
    4.45
    Should
    4.24
     must
    4.17
     должен
    3.95
     باید
    3.91
     harus
    3.88
     SHOULD
    3.84
     должны
    3.84
    Act Density 1.146%

    No Known Activations