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

    Neuronpedia

    INDEX
    Explanations

    propaganda and narrative control

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     misuse
    0.45
     ڈیٹ
    0.42
     inappropriately
    0.40
     verme
    0.40
     indiscrimin
    0.39
     inm
    0.38
    不起
    0.38
     unpredictable
    0.38
     agon
    0.38
     inappropri
    0.38
    POSITIVE LOGITS
     propaganda
    1.25
     Propaganda
    0.97
    宣传
    0.96
     propagand
    0.96
     narrative
    0.94
    propag
    0.86
     narrativa
    0.84
     narratives
    0.82
     Narrative
    0.82
     propagated
    0.79
    Act Density 0.036%

    No Known Activations