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

    Neuronpedia

    INDEX
    Explanations

    efficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    m
    1.01
    тся
    0.88
     graced
    0.86
     assorted
    0.85
     Bapak
    0.83
     л
    0.83
    ेंज
    0.82
    ↵
    0.80
    féle
    0.79
     woes
    0.78
    POSITIVE LOGITS
    *
    0.87
     banknotes
    0.82
    おしゃれ
    0.82
    が良い
    0.82
     bolts
    0.80
     तरीकों
    0.77
    れた
    0.76
    .!
    0.76
     شهر
    0.75
    ining
    0.75
    Act Density 4.796%

    No Known Activations