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

    Neuronpedia

    INDEX
    Explanations

    crime statistically

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ées
    1.43
    IBLE
    1.30
    ibility
    1.29
    АС
    1.28
    ี
    1.24
     métallique
    1.23
     fabriqué
    1.23
     rappel
    1.20
     fabriqu
    1.20
     имеются
    1.20
    POSITIVE LOGITS
    א
    1.32
    вая
    1.27
    此之外
    1.23
     reinvent
    1.21
     Attempts
    1.16
    стан
    1.11
    iye
    1.11
    iens
    1.11
    شة
    1.10
     ambito
    1.10
    Act Density 0.013%

    No Known Activations