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

    Neuronpedia

    INDEX
    Explanations

    Snowflake comparison

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    謝
    0.46
     ලි
    0.38
     Nig
    0.37
     Лон
    0.37
     calculadora
    0.37
     dispense
    0.36
    叟
    0.36
    тана
    0.36
    خته
    0.35
    nig
    0.35
    POSITIVE LOGITS
    warning
    0.45
     prompting
    0.42
    提示
    0.42
     warning
    0.41
    حظ
    0.40
     deserving
    0.39
    DC
    0.38
    警告
    0.38
    웜
    0.38
     +-
    0.38
    Act Density 0.002%

    No Known Activations