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

    Neuronpedia

    INDEX
    Explanations

    specific labels followed by value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Clar
    0.45
    פי
    0.40
     penggunaan
    0.40
     prestigio
    0.39
    피
    0.39
    最后的
    0.39
    сно
    0.38
    Family
    0.38
    Shirt
    0.38
    стка
    0.38
    POSITIVE LOGITS
     ஆயிரம்
    0.46
     poz
    0.42
    yl
    0.42
    atars
    0.41
     travaille
    0.41
     décou
    0.41
     accompanying
    0.41
     às
    0.40
     jeżeli
    0.40
     sous
    0.40
    Act Density 0.003%

    No Known Activations