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

    Neuronpedia

    INDEX
    Explanations

    fantasy worlds and genres

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ROC
    1.06
    Rocks
    1.02
     Kremlin
    0.98
    ні
    0.97
    서는
    0.97
    ной
    0.95
    TopOf
    0.94
     LEC
    0.93
     POC
    0.92
    ता
    0.90
    POSITIVE LOGITS
    ে
    1.19
    i
    1.18
    aan
    1.16
    aaa
    1.15
     entro
    1.13
    ेट्टी
    1.13
    ഹ്ലാദ
    1.10
     deber
    1.09
    aient
    1.09
    aa
    1.08
    Act Density 0.063%

    No Known Activations