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

    Neuronpedia

    INDEX
    Explanations

    capital of countries or capital ship

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     постро
    0.78
    સ
    0.72
    یک
    0.63
     인한
    0.61
     ایسا
    0.60
     añad
    0.59
    тся
    0.58
    える
    0.58
    ある
    0.57
     algebras
    0.57
    POSITIVE LOGITS
    cap
    1.09
     capt
    0.98
     Cap
    0.91
     Capt
    0.91
     CAP
    0.89
    capt
    0.89
    Cap
    0.89
    CAP
    0.89
     कैप
    0.88
     cap
    0.86
    Act Density 0.041%

    No Known Activations