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

    Neuronpedia

    INDEX
    Explanations

    tube, tubes, diameter, rack

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ни
    1.54
    dı
    1.15
    ння
    1.09
    ิ
    1.07
    ین
    1.05
    time
    1.05
    st
    1.04
    ่
    1.04
    ק
    1.04
    daughter
    1.02
    POSITIVE LOGITS
    大
    1.57
     tubes
    1.56
    个
    1.42
    어
    1.30
    我
    1.27
     tube
    1.23
    一些
    1.22
    你
    1.20
     Tubes
    1.18
    وم
    1.10
    Act Density 0.002%

    No Known Activations