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

    Neuronpedia

    INDEX
    Explanations

    asking user for input

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵
    1.31
     OL
    1.16
     we
    1.12
     paradox
    1.11
     heard
    1.08
     JavaScript
    1.07
     realistic
    1.06
     want
    1.06
     расскажем
    1.05
     reproduction
    1.05
    POSITIVE LOGITS
    The
    1.79
    These
    1.68
    This
    1.62
    Please
    1.60
    Review
    1.55
    You
    1.54
    Choose
    1.52
    Then
    1.49
    Their
    1.48
    По
    1.46
    Act Density 0.192%

    No Known Activations