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

    Neuronpedia

    INDEX
    Explanations

    create, choose, or customize

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     correctly
    0.44
     succeeding
    0.41
     correctamente
    0.41
    適切
    0.41
    successful
    0.40
     korrekt
    0.39
    ئت
    0.38
     successful
    0.38
     deserves
    0.37
     правильно
    0.37
    POSITIVE LOGITS
     customize
    0.97
     customise
    0.92
     facilmente
    0.87
     easily
    0.84
     choose
    0.82
    选择
    0.80
    Customize
    0.77
     Customize
    0.76
     fácilmente
    0.75
     выбрать
    0.75
    Act Density 0.040%

    No Known Activations