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

    Neuronpedia

    INDEX
    Explanations

    refusal, denied, ignored

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Faster
    0.42
     Context
    0.41
     Sexy
    0.40
     Spa
    0.38
     Optimal
    0.38
    важа
    0.38
     Falling
    0.38
     Bewer
    0.38
     Expectations
    0.37
    养
    0.37
    POSITIVE LOGITS
     আশ্বাস
    0.81
     refusal
    0.70
     refused
    0.69
     refus
    0.68
     refuses
    0.66
     refusing
    0.66
     shrugged
    0.66
     told
    0.64
    promised
    0.64
     refuse
    0.63
    Act Density 0.007%

    No Known Activations