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

    Neuronpedia

    INDEX
    Explanations

    rapidly changing and evolving

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ni
    0.28
    ili
    0.26
    rian
    0.25
     shoes
    0.24
    0.24
     (
    0.23
    etr
    0.23
    ong
    0.22
    nu
    0.22
    ito
    0.22
    POSITIVE LOGITS
     tijekom
    0.27
    ת
    0.26
     tremendously
    0.25
    ر
    0.24
    <unused282>
    0.24
    榱
    0.24
     linearly
    0.23
     incompar
    0.23
    ز
    0.23
    quela
    0.23
    Act Density 0.162%

    No Known Activations