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

    Neuronpedia

    INDEX
    Explanations

    white noise

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    recorder
    0.54
    recorded
    0.46
     അദ്ദേഹ
    0.43
    袮
    0.43
    moder
    0.43
     apology
    0.43
     profite
    0.43
     senator
    0.42
     monsieur
    0.42
    chairman
    0.42
    POSITIVE LOGITS
    こんにちは
    0.45
     noise
    0.43
     noises
    0.43
     crashes
    0.41
    糙
    0.40
    Noise
    0.39
     Рассмотрим
    0.39
     Noise
    0.38
    噪声
    0.38
     noisy
    0.38
    Act Density 0.002%

    No Known Activations