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

    Neuronpedia

    INDEX
    Explanations

    cleaned data, assumptions, origin

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.42
    /
    0.36
     job
    0.32
     like
    0.32
     over
    0.31
     class
    0.31
    ↵↵
    0.31
     cross
    0.31
    SLAM
    0.31
     and
    0.31
    POSITIVE LOGITS
    銠
    0.34
    <unused619>
    0.34
     ඔබට
    0.34
    ٘
    0.33
    眙
    0.33
    <unused1642>
    0.33
    <unused424>
    0.33
    質な
    0.32
    <unused597>
    0.32
     औष
    0.32
    Act Density 0.000%

    No Known Activations