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

    Neuronpedia

    INDEX
    Explanations

    arc length and shapes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    n
    1.62
    el
    1.52
    it
    1.49
    at
    1.45
    é
    1.41
    -
    1.40
    is
    1.37
    ни
    1.34
    q
    1.33
    是
    1.30
    POSITIVE LOGITS
     arc
    1.44
     Arc
    1.35
     arcs
    1.13
     وکړئ
    1.05
     extinguish
    0.98
     arco
    0.97
     rač
    0.96
     archi
    0.95
     établir
    0.95
     exécut
    0.94
    Act Density 0.006%

    No Known Activations