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

    Neuronpedia

    INDEX
    Explanations

    recommended way

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    有可能
    0.46
     unlikely
    0.45
     jenis
    0.45
    jenis
    0.45
     نوع
    0.43
    㣱
    0.43
    形式
    0.43
     Jenis
    0.42
     احتمال
    0.42
    Jenis
    0.42
    POSITIVE LOGITS
     elegant
    0.70
     recommended
    0.69
    Recommended
    0.66
     Recommended
    0.66
    Elegant
    0.64
     neat
    0.64
     elegante
    0.63
    recommended
    0.63
     clean
    0.61
     cleanest
    0.60
    Act Density 0.106%

    No Known Activations