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

    Neuronpedia

    INDEX
    Explanations

    emphasized improvement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Và
    0.52
    Responsibilities
    0.49
    คุณ
    0.47
    Sunday
    0.46
    يمة
    0.45
    ս
    0.44
    贰
    0.44
    답니다
    0.43
     Abgerufen
    0.43
    Arsenal
    0.43
    POSITIVE LOGITS
     emphasized
    0.49
     rod
    0.49
     cartridge
    0.47
     nitrogen
    0.46
     torus
    0.46
     rods
    0.46
     deliberately
    0.45
     pellet
    0.45
     ray
    0.45
     intentionally
    0.45
    Act Density 0.010%

    No Known Activations