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

    Neuronpedia

    INDEX
    Explanations

    diagrams, lists, code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    แหน่ง
    0.46
     מדי
    0.41
    тары
    0.38
    Transit
    0.38
     Kanada
    0.38
    透露
    0.36
     Pennington
    0.36
     Transit
    0.36
    国立
    0.36
    Tatar
    0.36
    POSITIVE LOGITS
     ↑
    0.48
    colors
    0.40
    ︽
    0.37
    わりに
    0.37
     цветов
    0.36
    edk
    0.36
     думать
    0.35
     Includes
    0.34
     diminish
    0.34
     callous
    0.34
    Act Density 0.000%

    No Known Activations