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

    Neuronpedia

    INDEX
    Explanations

    assigning values or properties

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    1.35
    1.09
    ം
    1.02
    い
    0.99
    其他
    0.95
    U
    0.95
    al
    0.93
    ం
    0.92
    ات
    0.92
     and
    0.91
    POSITIVE LOGITS
     gerekli
    1.01
    리기
    0.97
    sgál
    0.94
     STADT
    0.94
    考えて
    0.93
    latego
    0.90
    тная
    0.89
    ┈┈
    0.89
     situazioni
    0.88
    必要な
    0.88
    Act Density 0.175%

    No Known Activations