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

    Neuronpedia

    INDEX
    Explanations

    been affected or compromised

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unequivocal
    0.31
    ق
    0.29
    ز
    0.28
     sombras
    0.27
     crucial
    0.27
    nu
    0.27
     norms
    0.27
     characterise
    0.27
     consider
    0.26
     monetization
    0.26
    POSITIVE LOGITS
     hebt
    0.28
     heeft
    0.27
    中有
    0.27
     Мето
    0.26
    )=
    0.26
    荑
    0.26
     হয়েছে
    0.25
     আহত
    0.25
     *)(
    0.25
     中
    0.25
    Act Density 0.157%

    No Known Activations