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

    Neuronpedia

    INDEX
    Explanations

    say arrows

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ></
    0.62
     Alaskan
    0.61
    একজন
    0.61
    >"
    0.61
    >'
    0.61
    >%
    0.60
    なくなる
    0.60
    >
    0.59
    >;
    0.59
    ()>
    0.59
    POSITIVE LOGITS
     $\
    1.28
    $\
    1.17
     à
    0.95
    à
    0.95
     →
    0.94
     &\
    0.87
     ->
    0.84
    Û
    0.79
     ----
    0.78
     -->
    0.78
    Act Density 0.132%

    No Known Activations