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

    Neuronpedia

    INDEX
    Explanations

    typename

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Rear
    -0.07
     implication
    -0.07
    /+
    -0.06
     ACCESS
    -0.06
     intention
    -0.06
    ホ
    -0.06
     telling
    -0.06
     Sue
    -0.06
    ﭱ
    -0.06
    credit
    -0.06
    POSITIVE LOGITS
    镐
    0.07
    村庄
    0.07
     içinde
    0.07
    工作者
    0.07
    ırl
    0.07
    slideDown
    0.07
     stan
    0.07
    웃
    0.06
    �
    0.06
    ับ
    0.06
    Act Density 0.008%

    No Known Activations