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

    Neuronpedia

    INDEX
    Explanations

    reduction or decrease

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    นี้
    0.40
     domain
    0.40
    </h2>
    0.39
     Retriever
    0.38
    This
    0.38
     tractor
    0.38
    ک
    0.38
     triathlon
    0.37
     retriever
    0.37
     Jensen
    0.36
    POSITIVE LOGITS
     rédu
    0.42
     режима
    0.39
    reduce
    0.37
    en
    0.36
    ed
    0.36
     değişik
    0.35
     réduit
    0.35
     aé
    0.35
    on
    0.34
     réduire
    0.34
    Act Density 0.222%

    No Known Activations