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

    Neuronpedia

    INDEX
    Explanations

    conjunctions and separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ją
    0.41
     чувства
    0.40
    卻
    0.38
    珪
    0.37
    НЫ
    0.37
    ñar
    0.37
     состояния
    0.36
     तिथि
    0.35
     شدن
    0.35
    кратно
    0.35
    POSITIVE LOGITS
    Amp
    0.49
     doppia
    0.42
     subsidiary
    0.41
     T
    0.40
    matic
    0.39
     northern
    0.38
     foremost
    0.38
     P
    0.38
     twin
    0.38
     affiliated
    0.38
    Act Density 0.016%

    No Known Activations