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

    Neuronpedia

    INDEX
    Explanations

    pros, cons, no

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     emplacement
    0.45
    ٹ
    0.43
    Receive
    0.42
    であり
    0.41
    ზი
    0.41
     جیس
    0.40
    姶
    0.40
     quality
    0.40
     masterpiece
    0.40
     repetition
    0.39
    POSITIVE LOGITS
     δε
    0.43
    မဟုတ်
    0.42
     Onun
    0.42
     alternativ
    0.42
    ؟
    0.42
     pertanyaan
    0.41
    ？”
    0.41
     quelle
    0.40
     sonuc
    0.40
     Maybe
    0.40
    Act Density 0.002%

    No Known Activations