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

    Neuronpedia

    INDEX
    Explanations

    actions leading to outcomes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     unmistakable
    0.45
     inevitably
    0.41
    Ministry
    0.40
    ्राष्ट
    0.40
     discarded
    0.39
     invariably
    0.39
     domestic
    0.38
     hardly
    0.38
     unintended
    0.38
    Domestic
    0.38
    POSITIVE LOGITS
    maybe
    0.43
     pravidel
    0.43
     Maybe
    0.43
     misschien
    0.42
     Flexible
    0.42
     interrump
    0.42
     Flexibility
    0.40
    的空间
    0.40
    かもしれませんが
    0.39
     Struct
    0.39
    Act Density 0.009%

    No Known Activations