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

    Neuronpedia

    INDEX
    Explanations

    best practice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    etzte
    1.29
    น
    1.28
    ع
    1.20
    men
    1.19
    aden
    1.14
    lx
    1.13
    SO
    1.11
    unfortunately
    1.11
    atten
    1.11
    er
    1.09
    POSITIVE LOGITS
     ಯಾವುದ
    1.16
    ogu
    1.11
     drunk
    1.11
     органів
    1.10
     kitchen
    1.10
    值的
    1.09
    1.06
     Arcadia
    1.05
     metamorph
    1.05
     Utica
    1.03
    Act Density 0.001%

    No Known Activations