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

    Neuronpedia

    INDEX
    Explanations

    absolute refusal

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     संभावित
    0.35
     нә
    0.34
     dàn
    0.34
    靠
    0.34
    就连
    0.33
    MgO
    0.33
    潜在
    0.32
    romagn
    0.32
    ছোট
    0.31
     chhoti
    0.31
    POSITIVE LOGITS
     refusal
    0.47
     refuses
    0.46
     отказыва
    0.45
     refuse
    0.44
     refus
    0.43
     отказа
    0.43
     refused
    0.41
     refusing
    0.41
     declines
    0.40
    取消
    0.40
    Act Density 0.039%

    No Known Activations