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

    Neuronpedia

    INDEX
    Explanations

    spaces, areas, sectors, landscape

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ра
    1.19
    ваны
    1.06
    一方で
    1.03
    ён
    0.93
    𝒹
    0.92
    ină
    0.92
    ched
    0.90
    unts
    0.88
     welche
    0.88
    uje
    0.87
    POSITIVE LOGITS
     мелдеш
    0.94
    ❯
    0.90
     suku
    0.89
    zelf
    0.89
     sizlerle
    0.85
     mga
    0.84
    faktor
    0.84
    father
    0.83
     galima
    0.83
    voorbeeld
    0.83
    Act Density 0.121%

    No Known Activations