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

    Neuronpedia

    INDEX
    Explanations

    say "single uppercase letter"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ่วง
    -0.07
    ]=(
    -0.07
    ￣｀
    -0.07
    ابد
    -0.06
    ALL
    -0.06
    oving
    -0.06
    ermo
    -0.06
    =./
    -0.06
    _#{
    -0.06
     [/
    -0.06
    POSITIVE LOGITS
     uncertainty
    0.07
     Married
    0.06
    rien
    0.06
     органи
    0.06
    Caps
    0.06
    척
    0.06
    \system
    0.06
     franch
    0.06
     soul
    0.06
     calling
    0.06
    Act Density 0.006%

    No Known Activations