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

    Neuronpedia

    INDEX
    Explanations

    focuses on initialization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     эр
    0.54
    ంట్
    0.52
     форми
    0.51
     ту
    0.46
     ер
    0.45
     стандарт
    0.45
     кине
    0.44
     იმ
    0.44
     ℝ
    0.43
     ایپل
    0.43
    POSITIVE LOGITS
    know
    0.53
    strange
    0.44
    pat
    0.44
     warga
    0.43
    page
    0.42
     Nodes
    0.42
    gro
    0.42
     know
    0.41
     conoz
    0.41
    citizens
    0.40
    Act Density 0.001%

    No Known Activations