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

    Neuronpedia

    INDEX
    Explanations

    intense and powerful descriptors

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    性质
    0.70
     Ideal
    0.69
     나누
    0.68
     тим
    0.65
    돕
    0.65
     Mitarbeiter
    0.64
     দিয়ে
    0.64
     Helper
    0.63
     Guidelines
    0.62
     পক্ষপাত
    0.62
    POSITIVE LOGITS
     formidable
    1.91
     menacing
    1.43
     unstoppable
    1.39
     ferocious
    1.38
     deadly
    1.35
     dangerous
    1.31
     tenacious
    1.29
     lethal
    1.28
     solide
    1.28
    dangerous
    1.28
    Act Density 0.256%

    No Known Activations