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

    Neuronpedia

    INDEX
    Explanations

    routing or rewriting

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     firsthand
    -0.08
     braz
    -0.07
     positi
    -0.07
    見る
    -0.07
     logist
    -0.07
    心得
    -0.07
    .mean
    -0.07
     dag
    -0.07
    ergic
    -0.07
     kant
    -0.07
    POSITIVE LOGITS
     默认
    0.09
    ?action
    0.09
    :NO
    0.08
    Dispatcher
    0.08
     большин
    0.08
    ğiz
    0.08
     encamin
    0.08
     Dispatcher
    0.08
     Middleware
    0.08
     routing
    0.08
    Act Density 0.003%

    No Known Activations