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

    Neuronpedia

    INDEX
    Explanations

    proposed methods and recent advancements

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    หรือ
    0.57
    正常
    0.55
    类的
    0.53
     totalité
    0.53
    ے
    0.53
     ваших
    0.53
     hoặc
    0.52
    ğin
    0.52
    或者
    0.51
     یا
    0.50
    POSITIVE LOGITS
     was
    0.57
     innovative
    0.57
     for
    0.52
     be
    0.52
    w
    0.52
     been
    0.52
    in
    0.48
    1
    0.48
    $\
    0.48
    S
    0.48
    Act Density 0.041%

    No Known Activations