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

    Neuronpedia

    INDEX
    Explanations

    instances of polite expressions or behaviors

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

    politeness and courteousness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    chargez
    -0.47
     Découvrez
    -0.39
     übersch
    -0.38
    Życiorys
    -0.38
     jesu
    -0.37
     Moscú
    -0.36
     Ministério
    -0.36
     Egipto
    -0.34
     Voyez
    -0.34
     bankası
    -0.34
    POSITIVE LOGITS
    polite
    2.00
     polite
    1.92
     Polite
    1.79
     politely
    1.59
     politeness
    1.39
     courteous
    1.08
     Rude
    0.93
     POLIT
    0.82
    Rude
    0.81
    Gentle
    0.79
    Act Density 0.003%

    No Known Activations