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

    Neuronpedia

    INDEX
    Explanations

    A

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     kf
    -0.07
     dealership
    -0.07
    媆
    -0.07
    ⟵
    -0.06
    歡迎
    -0.06
    缶
    -0.06
     Janeiro
    -0.06
    -scal
    -0.06
    深入
    -0.06
    쾀
    -0.06
    POSITIVE LOGITS
     II
    0.08
    ฮ
    0.07
    olum
    0.07
    ******
↵
    0.06
     sisters
    0.06
    Ctx
    0.06
     itemId
    0.06
    .function
    0.06
    附加
    0.06
    _funcs
    0.06
    Act Density 0.053%

    No Known Activations