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

    Neuronpedia

    INDEX
    Explanations

    respectful communication and boundaries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     algorithm
    0.91
     Diagnostics
    0.84
     diagnostic
    0.84
     analysis
    0.83
    分析
    0.82
     Diagnostic
    0.82
     analyses
    0.81
    算法
    0.80
     diagnostics
    0.80
     विश्लेषण
    0.80
    POSITIVE LOGITS
     respectful
    2.19
     etiquette
    2.15
     polite
    1.92
     courteous
    1.88
     manners
    1.88
     disrespectful
    1.85
     respeto
    1.72
    Respect
    1.69
     respect
    1.65
     politely
    1.60
    Act Density 0.906%

    No Known Activations