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

    Neuronpedia

    INDEX
    Explanations

    javadoc parameter and return tags

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     a
    1.08
     daje
    0.98
     vuole
    0.95
    t
    0.95
     Pesh
    0.93
     with
    0.93
    ן
    0.93
     Sancho
    0.91
     помогает
    0.91
     Lakeland
    0.91
    POSITIVE LOGITS
    ul
    1.10
    ud
    1.05
    ва
    1.03
    ص
    1.01
    ки
    0.98
    ub
    0.94
    其他
    0.93
    ض
    0.93
    ла
    0.92
    .
    0.92
    Act Density 0.076%

    No Known Activations