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

    Neuronpedia

    INDEX
    Explanations

    rarely seen diagnosed

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    тик
    0.41
     упомина
    0.40
    मीटर
    0.39
     interrom
    0.39
    MORD
    0.37
     উচ্ছ
    0.37
     упо
    0.36
     СМИ
    0.36
     Jazz
    0.36
     Чем
    0.36
    POSITIVE LOGITS
     mostly
    0.40
    ensively
    0.39
     المصفوفه
    0.39
    钎
    0.39
     Priority
    0.39
    ordable
    0.39
     margarine
    0.39
     shady
    0.38
    ေ့
    0.38
     mainly
    0.38
    Act Density 0.001%

    No Known Activations