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

    Neuronpedia

    INDEX
    Explanations

    beneath the surface

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     uży
    1.07
     senses
    1.06
    і
    1.06
    ｉ
    1.01
     может
    1.01
    ｙ
    1.01
     verwendet
    1.01
     puedes
    1.00
    👍
    1.00
     хорошее
    0.98
    POSITIVE LOGITS
     fraught
    1.20
     grizz
    1.05
     scathing
    1.03
     glimpses
    1.01
    Чер
    1.01
     prerog
    0.99
     shrouded
    0.99
    措置
    0.99
    끊
    0.98
     conglomerates
    0.98
    Act Density 0.238%

    No Known Activations