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

    Neuronpedia

    INDEX
    Explanations

    petting, cuddling, stroking, flirting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    s
    0.67
     Irish
    0.60
     Have
    0.59
    iteten
    0.59
     c
    0.59
     b
    0.59
    beit
    0.59
    anes
    0.57
    sberg
    0.57
     Year
    0.56
    POSITIVE LOGITS
    ин
    0.74
    ду
    0.70
    рабо
    0.68
     cuddling
    0.67
    在
    0.67
    </h2>
    0.66
    ین
    0.65
    وع
    0.65
    ছে
    0.64
    де
    0.64
    Act Density 0.023%

    No Known Activations