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

    Neuronpedia

    INDEX
    Explanations

    the followed by specific qualifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     가장
    0.43
     সবচেয়ে
    0.42
    最も
    0.41
     наиболее
    0.40
    的最
    0.39
     সবচেয়ে
    0.38
    最有
    0.36
    中最
    0.35
    变得
    0.34
     সবচে
    0.34
    POSITIVE LOGITS
     fact
    1.20
     Tatsache
    1.09
     lack
    1.03
     činjen
    0.86
     absence
    0.86
    fact
    0.82
     inability
    0.81
    lack
    0.81
     факт
    0.79
     faptul
    0.78
    Act Density 0.171%

    No Known Activations