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

    Neuronpedia

    INDEX
    Explanations

    square root logic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     thằng
    0.48
     arriba
    0.46
     golongan
    0.45
     alberga
    0.45
     theyre
    0.43
     vidas
    0.42
     almacenamiento
    0.42
     autoestima
    0.41
     ills
    0.40
     inmig
    0.40
    POSITIVE LOGITS
    自然
    0.45
     необы
    0.44
     surprising
    0.43
    ??
    0.43
    ????
    0.42
     ???
    0.41
    意
    0.41
    /?
    0.40
    Poke
    0.40
    ..?
    0.40
    Act Density 0.006%

    No Known Activations