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

    Neuronpedia

    INDEX
    Explanations

    stating or seeking knowledge

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    тена
    0.70
    osyl
    0.64
    最多的
    0.63
     বিশেষণ
    0.62
     دیتی
    0.61
     fascination
    0.60
    зовая
    0.59
    Impress
    0.58
    なかなか
    0.57
     दृ
    0.57
    POSITIVE LOGITS
     know
    4.74
     knows
    4.39
    know
    4.24
     Know
    4.19
    Know
    4.16
     knowing
    4.02
     knew
    3.99
     KNOW
    3.85
    知道
    3.82
     Knowing
    3.73
    Act Density 0.784%

    No Known Activations