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

    Neuronpedia

    INDEX
    Explanations

    The [Topic]

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Suitable
    0.79
     Haha
    0.79
     Между
    0.77
     Faites
    0.76
     Pero
    0.75
     Sim
    0.75
     وت
    0.75
     Otras
    0.74
     Biraz
    0.72
    とか
    0.72
    POSITIVE LOGITS
     문제는
    1.30
    군은
    1.22
     내용은
    1.21
    意思是
    1.18
    성은
    1.14
    之所以
    1.14
    들은
    1.13
     realize
    1.10
     sifat
    1.09
     수는
    1.09
    Act Density 1.013%

    No Known Activations