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

    Neuronpedia

    INDEX
    Explanations

    weaknesses, limits

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     routines
    0.63
     padrões
    0.61
    िक्सल
    0.59
    ographers
    0.58
    initions
    0.56
    िली
    0.56
    ipennis
    0.55
    igrams
    0.55
     pastime
    0.55
     whatnot
    0.55
    POSITIVE LOGITS
    !!!
    1.05
     THIS
    0.89
    !!!!!
    0.89
     VERY
    0.88
    THIS
    0.87
     AMAZING
    0.84
    !!!!
    0.84
     HUGE
    0.84
     absolutely
    0.83
     !!!
    0.83
    Act Density 0.001%

    No Known Activations