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

    Neuronpedia

    INDEX
    Explanations

    code includes and imports

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     head
    0.90
     heck
    0.80
     hell
    0.80
     pak
    0.79
     water
    0.78
     New
    0.77
     context
    0.75
     shot
    0.74
     Knight
    0.72
     rapid
    0.71
    POSITIVE LOGITS
    ifdef
    1.52
     noqa
    1.39
    include
    1.38
    endif
    1.25
    pragma
    1.21
    define
    1.17
    ifndef
    1.12
    ️⃣
    1.06
    nullable
    1.06
    ⃣
    1.05
    Act Density 0.038%

    No Known Activations