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

    Neuronpedia

    INDEX
    Explanations

    variable swapping/symmetry

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     voller
    -0.09
     urgently
    -0.08
    ail
    -0.08
     schließlich
    -0.08
    ,Object
    -0.08
     плод
    -0.08
     alley
    -0.07
    .Obj
    -0.07
    uelen
    -0.07
     (?)
    -0.07
    POSITIVE LOGITS
     verandert
    0.09
     изменение
    0.09
     ænd
    0.09
     altering
    0.08
    改变
    0.08
     تغيير
    0.08
     ugyan
    0.08
     alter
    0.08
     toggle
    0.07
     aumenta
    0.07
    Act Density 0.030%

    No Known Activations