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

    Neuronpedia

    INDEX
    Explanations

    names and specific structures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     loot
    0.43
     Noel
    0.42
     junior
    0.42
     Loot
    0.42
     censored
    0.41
     Greenway
    0.41
    Selector
    0.40
     naive
    0.40
     نصف
    0.40
     selector
    0.40
    POSITIVE LOGITS
     generalizes
    0.44
     pédicule
    0.42
     oscillates
    0.41
    ्
    0.40
     oscill
    0.39
     Acá
    0.39
     পরম্প
    0.38
     architectures
    0.38
     ক্ষেপণাস্ত্র
    0.36
    華
    0.36
    Act Density 0.003%

    No Known Activations