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

    Neuronpedia

    INDEX
    Explanations

    references to devices or systems that are powered or energized

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    powered by

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    AlterField
    -0.48
    SVN
    -0.45
    xFFFF
    -0.39
     예
    -0.39
    Banana
    -0.38
     question
    -0.38
    MENAFN
    -0.37
     JFrame
    -0.37
     kij
    -0.37
     difficult
    -0.37
    POSITIVE LOGITS
     Powered
    0.97
    Powered
    0.92
     powered
    0.88
    powered
    0.83
     powering
    0.79
     propOrder
    0.72
    powering
    0.71
     fueled
    0.71
     POWER
    0.70
    POWER
    0.69
    Act Density 0.015%

    No Known Activations