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

    Neuronpedia

    INDEX
    Explanations

    ZDF means regarding filter topology

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    К
    0.86
    𝗖
    0.85
    𝗯
    0.82
    𝐂
    0.77
     фай
    0.75
    torrent
    0.75
    𝗔
    0.74
    𝗣
    0.73
    Associated
    0.73
     valori
    0.71
    POSITIVE LOGITS
     vuelva
    0.98
     sôi
    0.84
     necesito
    0.83
     deepen
    0.82
     poner
    0.81
     amarillo
    0.81
     उसी
    0.80
     dłu
    0.80
     miembro
    0.79
    忙
    0.79
    Act Density 0.001%

    No Known Activations