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

    Neuronpedia

    INDEX
    Explanations

    explaining context and structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     for
    0.66
     by
    0.57
     from
    0.53
     Korea
    0.53
     the
    0.50
    個人
    0.47
     as
    0.46
     with
    0.46
     China
    0.46
     Stake
    0.46
    POSITIVE LOGITS
    reservoir
    0.48
    нун
    0.46
    fordert
    0.46
     സോ
    0.45
     اسلح
    0.44
    زالة
    0.44
     ನಾ
    0.44
    shooter
    0.43
     ঘাঁ
    0.43
    ِمض
    0.43
    Act Density 0.001%

    No Known Activations