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

    Neuronpedia

    INDEX
    Explanations

    exploit, abuse, or endanger

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    %。
    0.32
     οποία
    0.30
     Prozent
    0.29
     умолчанию
    0.29
    রিতে
    0.29
     δύο
    0.29
     দুটি
    0.29
    ள்
    0.29
     etcétera
    0.29
    /';
    0.28
    POSITIVE LOGITS
     and
    0.45
     และ
    0.41
    ,
    0.40
     आणि
    0.38
    ,,
    0.37
     ,
    0.34
     और
    0.33
     și
    0.33
    ،
    0.33
     그리고
    0.32
    Act Density 0.243%

    No Known Activations