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

    Neuronpedia

    INDEX
    Explanations

    ‌

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    inarily
    -0.10
     inimesed
    -0.09
     agbegbe
    -0.09
    ,you
    -0.09
     eniyan
    -0.09
     batla
    -0.09
     kojoj
    -0.09
    ేదు
    -0.09
     bekommst
    -0.09
     hittar
    -0.09
    POSITIVE LOGITS
    �
    0.08
    ‍
    0.08
    ↵↵
    0.08
    remaining
    0.08
    new
    0.08
    گذاری
    0.08
    ��
    0.08
    unit
    0.07
     mir
    0.07
    ‌
    0.07
    Act Density 0.006%

    No Known Activations