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

    Neuronpedia

    INDEX
    Explanations

    admittedlyrecommend

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     stupid
    0.42
     Stupid
    0.41
     because
    0.40
     ponieważ
    0.38
     forskjellige
    0.38
     கூறியதாவது
    0.38
     bagus
    0.37
     which
    0.37
    했고
    0.37
     protože
    0.37
    POSITIVE LOGITS
     нередко
    0.43
    ábamos
    0.39
     égard
    0.39
    徬
    0.39
    へと
    0.38
    послед
    0.38
     অনেকটাই
    0.38
     ഏറെ
    0.38
    那就
    0.38
    こちらも
    0.37
    Act Density 0.199%

    No Known Activations