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

    Neuronpedia

    INDEX
    Explanations

    often, usually, frequently

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     chcesz
    0.49
    ollip
    0.46
    你需要
    0.45
    ouette
    0.43
     muốn
    0.43
    ্রো
    0.43
     ইন্দু
    0.43
     arrondies
    0.42
     necesitan
    0.42
    libexec
    0.42
    POSITIVE LOGITS
     often
    0.92
     अक्सर
    0.89
     vaak
    0.85
     spesso
    0.84
     často
    0.82
     często
    0.82
    大多数
    0.81
     oftentimes
    0.80
     many
    0.79
     Often
    0.79
    Act Density 0.133%

    No Known Activations