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

    Neuronpedia

    INDEX
    Explanations

    thoughts of harm

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     toxin
    0.44
     allergen
    0.40
    assapi
    0.40
     Monster
    0.39
     Diet
    0.39
    Diet
    0.38
     diet
    0.38
     Harwell
    0.38
    슝
    0.38
     epigen
    0.37
    POSITIVE LOGITS
    工
    0.42
    धरी
    0.41
    udah
    0.40
    Ă
    0.39
    лизм
    0.39
     мя
    0.38
     Alfredo
    0.38
    ăn
    0.38
    udh
    0.38
    ("#{
    0.38
    Act Density 0.004%

    No Known Activations