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

    Neuronpedia

    INDEX
    Explanations

    composition and related words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵↵
    0.54
    د
    0.47
    ↵↵
    0.46
    า
    0.46
    ع
    0.46
     औ
    0.44
     asl
    0.43
    ↵↵↵↵
    0.42
    }};
    0.42
     distant
    0.41
    POSITIVE LOGITS
     kompon
    0.67
     компози
    0.67
     kom
    0.66
     Ком
    0.64
    Ком
    0.64
     composição
    0.64
    Composition
    0.63
     ком
    0.62
     composición
    0.62
    Kom
    0.60
    Act Density 0.039%

    No Known Activations