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

    Neuronpedia

    INDEX
    Explanations

    sitting and seating positions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     acceler
    0.47
     ›
    0.47
     accelerating
    0.46
     spinoff
    0.46
     irrational
    0.45
     reload
    0.45
     learn
    0.44
     healthcare
    0.44
     goals
    0.44
     errors
    0.44
    POSITIVE LOGITS
     seating
    1.09
     앉
    1.07
    坐
    1.01
    座位
    1.00
     seated
    0.98
     sitting
    0.94
    坐在
    0.94
     duduk
    0.94
    sitting
    0.94
    Seat
    0.93
    Act Density 0.090%

    No Known Activations