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

    Neuronpedia

    INDEX
    Explanations

    qualifier followed by term

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     engages
    0.40
     बुनिया
    0.39
     Pillars
    0.39
    usia
    0.38
     Grundlage
    0.38
     عمد
    0.38
     reproduces
    0.38
     pillars
    0.37
    pillars
    0.37
    pillar
    0.37
    POSITIVE LOGITS
     alın
    0.48
    ाल
    0.47
     surgical
    0.46
    手术
    0.46
    та
    0.45
    ﺼ
    0.44
    痙
    0.44
     quasi
    0.44
     Subject
    0.43
    ف
    0.43
    Act Density 0.007%

    No Known Activations