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

    Neuronpedia

    INDEX
    Explanations

    defining or structuring inputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ankaŭ
    0.34
     även
    0.32
     idha
    0.32
     följande
    0.30
     taea
    0.29
     tohoto
    0.29
     tôi
    0.29
     kobiet
    0.29
     Spicy
    0.29
     Barley
    0.29
    POSITIVE LOGITS
    way
    0.33
    determining
    0.32
     определяет
    0.32
    set
    0.30
    Съ
    0.30
    Ъ
    0.30
    which
    0.30
    пу
    0.29
     плюс
    0.29
    물의
    0.29
    Act Density 0.387%

    No Known Activations