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

    Neuronpedia

    INDEX
    Explanations

    add or integrate actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     complying
    0.24
     belongs
    0.23
     complies
    0.22
     elde
    0.21
    Destination
    0.21
                           
    0.20
    communication
    0.20
     Accessed
    0.20
     receives
    0.20
    pter
    0.20
    POSITIVE LOGITS
     добавить
    0.36
     постара
    0.33
     add
    0.31
     put
    0.31
     incorporate
    0.31
     попробовать
    0.31
     menambahkan
    0.30
    让
    0.30
     include
    0.29
     попыта
    0.29
    Act Density 0.196%

    No Known Activations