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

    Neuronpedia

    INDEX
    Explanations

    unwanted sexual or harassing behavior

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     masterpiece
    0.54
     plug
    0.52
     killer
    0.51
     trillions
    0.50
     doom
    0.49
     optimized
    0.49
     dynamically
    0.48
     civilizations
    0.46
     optimization
    0.46
     evils
    0.46
    POSITIVE LOGITS
     uncomfortable
    0.82
     harassing
    0.80
     harassment
    0.80
     intimidation
    0.78
     inappropriate
    0.77
     humiliating
    0.74
     escalating
    0.72
     conductas
    0.72
     comportamenti
    0.71
     discomfort
    0.70
    Act Density 0.034%

    No Known Activations