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

    Neuronpedia

    INDEX
    Explanations

    additional or special requirements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ที่ดี
    1.05
     lehető
    0.95
     suficiente
    0.94
     assuré
    0.94
     verantwoord
    0.93
     صحیح
    0.91
     clairement
    0.90
     appropriately
    0.89
     adecuado
    0.89
     suficient
    0.88
    POSITIVE LOGITS
     expensive
    1.28
     costly
    1.11
     special
    1.10
    额外
    1.10
    担心
    1.09
     manually
    1.08
    特別な
    1.06
     Expensive
    1.02
    额外的
    1.00
     additional
    0.97
    Act Density 0.108%

    No Known Activations