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

    Neuronpedia

    INDEX
    Explanations

    negative character traits and dangerous behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    evil, villain, menacing, ruthless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Atsauces
    -0.46
     kasarigan
    -0.43
     endwhile
    -0.42
    Personensuche
    -0.41
     blunder
    -0.41
     verständ
    -0.38
    addPreferredGap
    -0.38
    addGap
    -0.37
     @"/
    -0.37
    cstdio
    -0.37
    POSITIVE LOGITS
    😈
    0.48
     ruthless
    0.47
    الحياه
    0.47
     악
    0.44
     pleaſure
    0.44
     villain
    0.43
     coldly
    0.43
    TextInputLayout
    0.42
    狠狠
    0.42
    recated
    0.42
    Act Density 0.086%

    No Known Activations