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

    Neuronpedia

    INDEX
    Explanations

    farewells

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ।
    1.12
    。
    1.05
    1
    1.04
    På
    0.96
     powied
    0.92
    、
    0.89
    di
    0.88
    K
    0.88
    )");
    0.87
     {
    0.85
    POSITIVE LOGITS
    จะ
    0.96
    ை
    0.91
    ция
    0.89
    一個
    0.88
    тор
    0.87
     смерть
    0.87
     jeopard
    0.87
     ecstasy
    0.86
     byen
    0.86
     약
    0.85
    Act Density 0.001%

    No Known Activations