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

    Neuronpedia

    INDEX
    Explanations

    parameters and criteria determining actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     skewers
    0.43
     addObject
    0.40
    季度
    0.40
    जेट
    0.38
     ihrer
    0.38
     slat
    0.38
     smother
    0.37
     immobil
    0.37
     Saar
    0.37
     چې
    0.37
    POSITIVE LOGITS
    abilă
    0.52
    ाब
    0.50
    і
    0.50
    า
    0.50
    ל
    0.48
    =_
    0.47
    лі
    0.46
    ла
    0.46
     выполнить
    0.45
    ж
    0.45
    Act Density 0.002%

    No Known Activations