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

    Neuronpedia

    INDEX
    Explanations

    existing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    𝑖
    3.30
    いった
    3.30
    ্স
    3.28
    𝑢
    3.07
    𝑝
    3.02
    ার
    2.93
    со
    2.93
    ר
    2.85
    ন
    2.84
    써
    2.83
    POSITIVE LOGITS
    iciency
    2.83
    फारिश
    2.75
     decía
    2.40
    ென
    2.37
    ه
    2.35
     cevap
    2.35
     После
    2.34
     tespit
    2.30
    ূট
    2.30
     faithfully
    2.30
    Act Density 0.053%

    No Known Activations