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

    Neuronpedia

    INDEX
    Explanations

    constant attention

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    经纪
    0.49
     optimally
    0.45
     Untersuchung
    0.45
     успешно
    0.45
    <unused73>
    0.44
    ії
    0.43
     stratég
    0.43
    خدم
    0.42
     υπηρε
    0.42
    licz
    0.41
    POSITIVE LOGITS
    Runtime
    0.51
    Departure
    0.49
    rupole
    0.47
     compositional
    0.47
    Composition
    0.46
     बाती
    0.46
    Getter
    0.46
    Pakistan
    0.45
    getBounding
    0.45
    PathParam
    0.44
    Act Density 0.002%

    No Known Activations