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

    Neuronpedia

    INDEX
    Explanations

    static, rule, crucial

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    le
    0.46
    기
    0.46
     ECS
    0.45
     DCS
    0.45
    ging
    0.44
    brat
    0.44
    bots
    0.44
    quels
    0.44
    t
    0.44
    class
    0.43
    POSITIVE LOGITS
    ত্যাশিত
    0.51
    🧂
    0.50
     cayenne
    0.50
     mosque
    0.49
    Diam
    0.48
    🦳
    0.48
     semicircle
    0.47
     roadside
    0.47
     金属
    0.46
     clín
    0.46
    Act Density 0.141%

    No Known Activations