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

    Neuronpedia

    INDEX
    Explanations

    potential loss or directly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    REIB
    0.35
     Gabb
    0.34
    래
    0.34
    앙
    0.33
     Internet
    0.33
    COUNTRY
    0.33
     respectivas
    0.33
     Forma
    0.32
     Drucker
    0.32
    ქვ
    0.31
    POSITIVE LOGITS
    っていた
    0.36
    斦
    0.36
     شود
    0.35
     trotz
    0.35
    やり
    0.34
    ريكي
    0.34
    хой
    0.34
     zelfs
    0.34
    WhiteElo
    0.33
    ಲಾಯಿತು
    0.33
    Act Density 0.002%

    No Known Activations