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

    Neuronpedia

    INDEX
    Explanations

    clean and efficient

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     =
    1.31
    =
    1.27
     envis
    1.14
    ousal
    1.13
    ŵ
    1.08
    ial
    1.07
     Betting
    1.06
     =$
    1.05
     hin
    1.03
    ulating
    1.03
    POSITIVE LOGITS
     FRESH
    1.34
     pristine
    1.32
     cleanly
    1.31
     spotless
    1.30
    fresh
    1.28
     concise
    1.28
     uncomplicated
    1.26
     freshly
    1.24
     effortless
    1.23
     sauber
    1.23
    Act Density 0.369%

    No Known Activations