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

    Neuronpedia

    INDEX
    Explanations

    files documents

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    contents
    0.99
    我想
    0.89
    o
    0.88
     tính
    0.84
     disapp
    0.83
    陵
    0.83
    a
    0.80
    の内容
    0.80
    ",
    0.78
     dụng
    0.78
    POSITIVE LOGITS
    ثال
    1.30
     flirt
    1.12
     bombing
    1.10
    BrN
    1.09
     punching
    1.08
     steadfast
    1.08
    🆈
    1.07
    ભાઇ
    1.07
    āna
    1.06
    ັ້ງ
    1.06
    Act Density 0.000%

    No Known Activations