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

    Neuronpedia

    INDEX
    Explanations

    Jelqing or tokenization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    狨
    0.31
     Relativity
    0.29
     muziek
    0.29
     جھنڈ
    0.29
     Lakeside
    0.29
     موسیقی
    0.28
     قابلیت
    0.28
    经销商
    0.28
    маты
    0.27
     ليس
    0.27
    POSITIVE LOGITS
    how
    0.38
    Como
    0.33
    what
    0.31
    when
    0.31
    eden
    0.31
    ef
    0.31
    which
    0.30
     safest
    0.30
    How
    0.30
    Best
    0.30
    Act Density 0.000%

    No Known Activations