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

    Neuronpedia

    INDEX
    Explanations

    the core idea

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     importantes
    0.45
    这点
    0.42
    这一点
    0.41
     fact
    0.40
     wichtiger
    0.39
     Tatsache
    0.39
     importanti
    0.39
     najbardziej
    0.38
     mencionar
    0.38
     mattered
    0.38
    POSITIVE LOGITS
     premise
    1.27
     gist
    1.16
     basic
    0.98
    basic
    0.88
    基本
    0.88
    基本的な
    0.86
    的基本
    0.85
     idea
    0.85
     basis
    0.82
     essence
    0.79
    Act Density 0.082%

    No Known Activations