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

    Neuronpedia

    INDEX
    Explanations

    deal expert

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    并不是
    0.90
     પણ
    0.89
    这种
    0.85
     أيضا
    0.85
    這種
    0.84
     있지만
    0.84
     أيضاً
    0.84
     چنین
    0.84
     meantime
    0.84
     també
    0.84
    POSITIVE LOGITS
     a
    1.10
    ة
    0.98
    ik
    0.93
    ovací
    0.93
    el
    0.92
    id
    0.91
    लरशिप
    0.89
     an
    0.88
    omechanics
    0.88
    ငန်း
    0.88
    Act Density 0.013%

    No Known Activations