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

    Neuronpedia

    INDEX
    Explanations

    refusing harmful requests

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    agana
    0.81
     Nia
    0.78
     Вол
    0.76
     Castle
    0.74
     nai
    0.73
    还原
    0.72
     ഉത്തര
    0.71
     ductile
    0.71
     Zeiten
    0.70
     Vaill
    0.70
    POSITIVE LOGITS
    getWorld
    0.69
     следу
    0.67
     avoid
    0.66
     heap
    0.64
     ByteBuffer
    0.64
     кар
    0.64
     adhere
    0.62
     herds
    0.62
    র্ধ
    0.61
    мет
    0.60
    Act Density 0.044%

    No Known Activations