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

    Neuronpedia

    INDEX
    Explanations

    ends of sentences

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     decisions
    0.51
    这些
    0.50
     vielfält
    0.50
    เหล่านี้
    0.49
    這些
    0.48
     यामुळे
    0.47
     these
    0.46
     helps
    0.46
     enables
    0.46
     هذه
    0.46
    POSITIVE LOGITS
    }$).
    0.43
     కానీ
    0.43
    }$)
    0.41
    "]="
    0.41
    }$(
    0.41
    then
    0.40
    ]='
    0.40
     먼저
    0.40
    щую
    0.39
     তারপর
    0.38
    Act Density 0.416%

    No Known Activations