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

    Neuronpedia

    INDEX
    Explanations

    Multilingual terms for concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ים
    3.12
     twee
    2.72
    aar
    2.65
    eper
    2.64
     doar
    2.56
     trebui
    2.52
    ร์
    2.50
    hitung
    2.48
    ppery
    2.45
    ್‌
    2.44
    POSITIVE LOGITS
    nya
    4.83
    ো
    3.64
    نا
    3.45
    swith
    3.43
    م
    3.40
    ها
    3.37
    یت
    3.26
    n
    3.21
    ش
    3.21
    ことができる
    3.20
    Act Density 0.016%

    No Known Activations