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

    Neuronpedia

    INDEX
    Explanations

    direction of movement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Up
    1.08
     Up
    1.01
    UpDown
    0.97
    上げた
    0.87
    アップ
    0.83
    ড
    0.83
    upload
    0.82
    zték
    0.81
    いない
    0.81
     up
    0.80
    POSITIVE LOGITS
    wards
    1.44
    ward
    1.22
     into
    1.20
     wards
    1.10
    WARDS
    1.04
     towards
    1.02
    WARD
    0.98
     toward
    0.96
     onto
    0.96
     ward
    0.92
    Act Density 0.100%

    No Known Activations