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

    Neuronpedia

    INDEX
    Explanations

    many, much, few

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    lás
    0.72
     warto
    0.70
    eraj
    0.69
     રીતે
    0.68
     So
    0.68
     جاء
    0.67
     أت
    0.66
     Ona
    0.65
    лла
    0.64
    ம
    0.64
    POSITIVE LOGITS
     много
    1.79
     nhiều
    1.76
     הרבה
    1.74
     несколько
    1.74
     banyak
    1.67
     mnogo
    1.63
     quelques
    1.61
     багато
    1.56
     beaucoup
    1.56
     кілька
    1.55
    Act Density 0.069%

    No Known Activations