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

    Neuronpedia

    INDEX
    Explanations

    need, will, what, should

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     પરંતુ
    0.34
     כאשר
    0.31
    然而
    0.30
     然而
    0.30
     परंतु
    0.29
     Embora
    0.29
     ولكن
    0.28
    Although
    0.27
     achieving
    0.27
     ಆದ್ದರಿಂದ
    0.27
    POSITIVE LOGITS
     magari
    0.37
     rethink
    0.35
    гант
    0.33
    看看
    0.32
    бл
    0.31
     какой
    0.30
    ویی
    0.30
     нужен
    0.30
    買う
    0.30
    люми
    0.30
    Act Density 0.072%

    No Known Activations