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

    Neuronpedia

    INDEX
    Explanations

    midpoint calculation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    k
    1.75
    ant
    1.59
    t
    1.55
    c
    1.31
    ov
    1.28
    et
    1.27
    el
    1.26
    dır
    1.26
    1
    1.25
    tenths
    1.24
    POSITIVE LOGITS
     До
    1.55
     ООО
    1.26
    Ა
    1.22
     ويمكن
    1.20
     ਡ
    1.20
    ⚉
    1.20
     Ο
    1.18
    1.17
     도
    1.16
    𝓐
    1.16
    Act Density 0.016%

    No Known Activations