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

    Neuronpedia

    INDEX
    Explanations

    feeding back outputs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     therefrom
    0.45
    焉
    0.44
    від
    0.44
     oryginal
    0.43
    чнай
    0.43
    ⃘
    0.40
    ซีน
    0.40
    苖
    0.39
    Balances
    0.39
    🥑
    0.39
    POSITIVE LOGITS
    加上
    0.54
     ಸೇ
    0.44
     adicion
    0.42
    再加上
    0.41
     performed
    0.39
     add
    0.39
     added
    0.39
     cum
    0.38
     إضافة
    0.38
    添加到
    0.38
    Act Density 0.001%

    No Known Activations