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

    Neuronpedia

    INDEX
    Explanations

    upfront prevents

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     reuniones
    0.47
    واز
    0.41
    زاب
    0.39
    زام
    0.36
    会议
    0.35
     внутри
    0.35
     shortly
    0.35
    婚
    0.34
     risultati
    0.34
    报警
    0.33
    POSITIVE LOGITS
     automobil
    0.41
     llor
    0.41
     verhindert
    0.41
     ag
    0.38
     proxies
    0.38
     afraid
    0.38
     kors
    0.38
    ূল্য
    0.38
     utmost
    0.37
    Um
    0.37
    Act Density 0.001%

    No Known Activations