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

    Neuronpedia

    INDEX
    Explanations

    available 24/7

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ꭸ
    0.41
    悵
    0.38
     progressivement
    0.37
     LOTRAchievement
    0.36
    ურთიერთ
    0.36
    ettin
    0.35
    Periodic
    0.35
    优雅
    0.34
    逐渐
    0.34
    rements
    0.34
    POSITIVE LOGITS
     round
    1.37
     nonstop
    1.31
     २४
    1.29
     constant
    1.18
    round
    1.14
    ตลอด
    1.09
     Round
    1.08
     continuous
    1.07
    Round
    1.07
     anytime
    1.06
    Act Density 0.059%

    No Known Activations