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

    Neuronpedia

    INDEX
    Explanations

    values and dates

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    🫓
    -1.52
     mergul
    -1.45
    茈
    -1.32
    鸰
    -1.30
     soportar
    -1.28
     teorías
    -1.27
     préparé
    -1.27
     MUCH
    -1.27
     apagar
    -1.25
     pudesse
    -1.25
    POSITIVE LOGITS
     if
    2.44
     or
    2.05
     for
    2.02
     any
    1.63
     unless
    1.55
     если
    1.52
     must
    1.45
     should
    1.42
     with
    1.39
    Если
    1.32
    Act Density 0.003%

    No Known Activations