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

    Neuronpedia

    INDEX
    Explanations

    Thai and Vietnamese "do," plus "pity"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    委员
    3.27
    ので
    3.17
    አ
    3.02
    お
    3.00
    い
    2.95
    क
    2.86
    员
    2.85
    ी
    2.85
     nhìn
    2.82
    en
    2.81
    POSITIVE LOGITS
    s
    4.72
    ्स
    3.79
    ের
    3.64
    ING
    3.63
    tion
    3.50
    tional
    3.41
    ta
    3.28
    ্টর
    3.27
    ই
    3.26
    tions
    3.22
    Act Density 0.014%

    No Known Activations