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

    Neuronpedia

    INDEX
    Explanations

    asking what you want

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     numerous
    0.61
     seamlessly
    0.60
     vastly
    0.60
     ඉතා
    0.59
     even
    0.58
     fully
    0.57
     även
    0.56
     whenever
    0.56
     starved
    0.56
     interchangeably
    0.55
    POSITIVE LOGITS
     тебе
    0.89
     тебя
    0.80
    你想
    0.72
     você
    0.71
    你的
    0.71
     তোমার
    0.70
     당신
    0.69
     тобі
    0.68
     necesitas
    0.68
     你
    0.67
    Act Density 0.095%

    No Known Activations