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

    Neuronpedia

    INDEX
    Explanations

    instant recognition/impact

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    -
    0.46
     "
    0.40
     reeds
    0.37
     sewer
    0.36
    ancock
    0.35
    ")
    0.35
     trac
    0.35
     Sawyer
    0.35
    salted
    0.35
    inerary
    0.34
    POSITIVE LOGITS
    ➀
    0.50
    흙
    0.44
     topLine
    0.44
    Фі
    0.43
    捎
    0.42
     ಜೀವನ
    0.41
    ;//
    0.41
    otherArchive
    0.41
    ,//
    0.40
    branco
    0.39
    Act Density 0.001%

    No Known Activations