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

    Neuronpedia

    INDEX
    Explanations

    military units and roles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     targeted
    0.89
     poisoned
    0.88
     attacks
    0.86
    古屋
    0.83
    attacks
    0.82
     Targeted
    0.81
     Elena
    0.81
     attaques
    0.81
     sabotage
    0.81
    targeted
    0.80
    POSITIVE LOGITS
     military
    1.87
    Military
    1.84
     Army
    1.84
     Military
    1.79
    Army
    1.68
    military
    1.66
     army
    1.65
    军
    1.58
     soldiers
    1.58
    軍
    1.54
    Act Density 0.389%

    No Known Activations