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

    Neuronpedia

    INDEX
    Explanations

    physics concepts (energy, pressure, heat)

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    น
    0.50
    ен
    0.48
    er
    0.45
     Վ
    0.44
    ز
    0.43
    な
    0.43
    。”
    0.43
    ੇ
    0.42
    तया
    0.42
    ینا
    0.42
    POSITIVE LOGITS
     volts
    0.49
     энерги
    0.43
    .
    0.42
    >
    0.42
    ?
    0.41
     ऊर्जा
    0.41
    8
    0.40
     currents
    0.39
     Energie
    0.39
    력을
    0.39
    Act Density 0.397%

    No Known Activations