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

    Neuronpedia

    INDEX
    Explanations

    put ' to' or similar structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tragen
    1.00
     construido
    0.91
     zumindest
    0.86
     costruito
    0.86
     zonder
    0.83
     múltiplos
    0.83
     ohne
    0.82
     Poké
    0.81
     gebaut
    0.79
     vollständig
    0.79
    POSITIVE LOGITS
    е
    1.09
    о
    0.92
    ет
    0.76
    ri
    0.75
    ні
    0.72
    r
    0.72
    го
    0.72
    ти
    0.70
    mi
    0.69
    Совет
    0.69
    Act Density 0.003%

    No Known Activations