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

    Neuronpedia

    INDEX
    Explanations

    strictly from, opens at, image shares

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.57
     L
    0.55
    j
    0.52
     I
    0.50
     Re
    0.49
    år
    0.49
    L
    0.48
    hed
    0.48
    za
    0.48
     E
    0.47
    POSITIVE LOGITS
    !!”
    0.57
    ！”
    0.56
    ્સ
    0.52
    !“
    0.52
    ᚑ
    0.51
     Tumblr
    0.51
    ,/*
    0.50
    !!!"
    0.50
    ，”
    0.49
     marchés
    0.48
    Act Density 0.000%

    No Known Activations