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

    Neuronpedia

    INDEX
    Explanations

    vibrant, complex

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     jeopardy
    0.59
    عى
    0.58
    меры
    0.57
     leaks
    0.55
     pasti
    0.55
    ции
    0.55
     catastrophes
    0.54
     casualties
    0.54
    いろ
    0.54
     megy
    0.54
    POSITIVE LOGITS
    ly
    0.91
    scient
    0.81
    amente
    0.68
    the
    0.67
    ‌ترین
    0.67
     ترین
    0.65
    fully
    0.64
    ll
    0.62
    multif
    0.61
    LY
    0.61
    Act Density 2.286%

    No Known Activations