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

    Neuronpedia

    INDEX
    Explanations

    dominant

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    。
    1.35
    。「
    1.09
    ת
    1.06
    ü
    1.02
    。《
    0.99
    دين
    0.98
     teléfono
    0.93
    。）
    0.93
    า
    0.93
    ።
    0.93
    POSITIVE LOGITS
     adjudged
    1.09
     incisions
    1.03
    αν
    1.02
    gleichen
    1.01
    ants
    0.96
     operands
    0.96
    the
    0.95
    ts
    0.95
    전
    0.94
     autochtones
    0.94
    Act Density 0.004%

    No Known Activations