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

    Neuronpedia

    INDEX
    Explanations

    stable training

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     прожи
    0.45
    isati
    0.41
    ଙ୍କ
    0.40
     Styles
    0.40
    ቱም
    0.40
     熱
    0.39
     Пан
    0.39
     criminally
    0.39
     Scissors
    0.38
    無料
    0.37
    POSITIVE LOGITS
    摇头
    0.41
    örungen
    0.38
    ammam
    0.38
     reag
    0.37
    0.37
    ولة
    0.35
    前半
    0.35
     دوربین
    0.35
     gorges
    0.34
     mayoral
    0.34
    Act Density 0.001%

    No Known Activations