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

    Neuronpedia

    INDEX
    Explanations

    method extensions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    alamualaikum
    -1.09
    marzo
    -1.04
     materiałów
    -1.02
     czter
    -1.02
     stří
    -1.00
    technologies
    -1.00
    KURZBESCHREIBUNG
    -1.00
     dvě
    -0.99
    我喜欢
    -0.98
     クッキー
    -0.98
    POSITIVE LOGITS
     to
    1.31
     how
    1.27
    5
    1.10
    9
    1.08
     adds
    1.06
    本身
    1.06
    、
    1.05
     such
    1.05
     Because
    1.05
    ريط
    1.01
    Act Density 0.013%

    No Known Activations