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

    Neuronpedia

    INDEX
    Explanations

    decoration and structure descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ;
    0.62
    q
    0.62
    )
    0.59
    1
    0.55
    :
    0.53
    is
    0.50
    se
    0.50
    9
    0.50
    ]
    0.48
    of
    0.46
    POSITIVE LOGITS
    itettura
    0.58
     dekor
    0.54
    花园
    0.52
    atelle
    0.50
     fleurs
    0.50
     아름
    0.50
    ेंट
    0.50
    iatura
    0.50
     bude
    0.49
     edificios
    0.48
    Act Density 0.163%

    No Known Activations