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

    Neuronpedia

    INDEX
    Explanations

    defending titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    conectar
    -0.81
     trợ
    -0.79
    risch
    -0.78
    この日は
    -0.77
    waarde
    -0.77
    ȵ
    -0.77
     Feuchtigkeit
    -0.75
    čkou
    -0.73
    expandafter
    -0.72
     permisos
    -0.72
    POSITIVE LOGITS
     defend
    4.19
     defending
    4.13
     defended
    3.61
     defends
    3.56
     Defend
    2.92
     defense
    2.81
     defender
    2.80
     defenders
    2.73
     defence
    2.70
     DEFEND
    2.44
    Act Density 0.059%

    No Known Activations