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

    Neuronpedia

    INDEX
    Explanations

    identifying options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     عادة
    0.70
    च्छ
    0.69
     observes
    0.67
     cần
    0.67
     requires
    0.67
     amacı
    0.66
     zwy
    0.65
     harmless
    0.65
     estrict
    0.64
     benötigt
    0.64
    POSITIVE LOGITS
     what
    1.49
    what
    1.21
     возможности
    1.11
     posibilidades
    1.10
    怎么样
    1.07
     mogelijkheden
    1.06
     whatever
    1.05
     результа
    1.04
     how
    1.04
     possibilità
    1.03
    Act Density 0.003%

    No Known Activations