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

    Neuronpedia

    INDEX
    Explanations

    spatial areas

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Martin
    0.42
     ориенти
    0.40
     مان
    0.38
    平成
    0.38
     line
    0.37
    line
    0.36
     wasting
    0.35
    gloss
    0.35
    strongly
    0.35
     atraer
    0.35
    POSITIVE LOGITS
    Box
    0.53
     Box
    0.50
     Boxes
    0.50
    盒
    0.49
     boxes
    0.48
    Area
    0.48
     Area
    0.47
     箱
    0.47
     box
    0.46
     AREA
    0.46
    Act Density 0.003%

    No Known Activations