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

    Neuronpedia

    INDEX
    Explanations

    superstition and history

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Брита
    0.45
     који
    0.45
     revamped
    0.44
    !!)
    0.44
    ances
    0.42
     ако
    0.42
    బ్యా
    0.42
    ដើម្បី
    0.42
     jotka
    0.42
     које
    0.42
    POSITIVE LOGITS
     höchste
    0.48
    A
    0.45
    并通过
    0.45
    p
    0.44
    рыв
    0.41
     ancak
    0.40
     leichter
    0.40
     TabLayout
    0.40
     byd
    0.38
    否则
    0.38
    Act Density 0.009%

    No Known Activations