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

    Neuronpedia

    INDEX
    Explanations

    code definitions or features

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Можно
    0.47
     گی۔
    0.45
    marshal
    0.43
     можно
    0.41
    Чтобы
    0.40
    chées
    0.40
    ंगन
    0.39
    olines
    0.39
    。
    0.39
    uyor
    0.39
    POSITIVE LOGITS
     du
    0.47
    ST
    0.44
     problems
    0.43
     hodin
    0.43
    啤
    0.43
    フレー
    0.42
     njia
    0.42
     problèmes
    0.42
     शुक्र
    0.41
     établ
    0.41
    Act Density 0.000%

    No Known Activations