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

    Neuronpedia

    INDEX
    Explanations

    any followed by specifics

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    an
    1.27
    いる
    1.25
    speople
    1.21
     eing
    1.20
     للغاية
    1.17
     things
    1.14
    мес
    1.09
    ใหญ
    1.09
    s
    1.07
     каждый
    1.07
    POSITIVE LOGITS
    wheres
    2.02
     kind
    1.81
    thin
    1.66
     combination
    1.63
     semblance
    1.61
    where
    1.52
    які
    1.52
    ق
    1.50
    cubic
    1.40
    hoo
    1.34
    Act Density 0.081%

    No Known Activations