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

    Neuronpedia

    INDEX
    Explanations

    using System.Linq

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    始终
    1.54
    BTW
    1.50
    ר
    1.47
    volatile
    1.42
    сть
    1.42
    Ethnic
    1.40
    除此之外
    1.40
    در
    1.39
    perceptron
    1.38
    து
    1.38
    POSITIVE LOGITS
     kada
    2.01
    ו
    1.90
    ির
    1.72
     Австра
    1.69
     parfait
    1.67
     Easier
    1.59
     $"{
    1.59
    тия
    1.57
    hive
    1.57
     tomando
    1.56
    Act Density 0.067%

    No Known Activations