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

    Neuronpedia

    INDEX
    Explanations

    measurements in inches

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ي
    0.56
    i
    0.52
    ة
    0.48
    י
    0.47
     of
    0.46
    ing
    0.46
    the
    0.42
    en
    0.39
     épouse
    0.39
    er
    0.38
    POSITIVE LOGITS
     moratorium
    0.32
    ни
    0.32
     sensit
    0.32
     spieg
    0.31
    赦
    0.30
    নাশ
    0.30
     chunky
    0.30
    ضب
    0.30
    िक्स
    0.29
    uglify
    0.29
    Act Density 0.002%

    No Known Activations