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

    Neuronpedia

    INDEX
    Explanations

    comparisons and arithmetic operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ({\
    0.23
    ={\
    0.22
    +{\
    0.21
     ({\
    0.20
    «.
    0.20
    ‌.
    0.20
    «,
    0.19
    ภัณฑ์
    0.19
    }$;
    0.19
    $=\
    0.19
    POSITIVE LOGITS
     %
    0.18
     정
    0.18
    <unused2136>
    0.17
     params
    0.17
     vacaciones
    0.17
     ~/
    0.17
     !=
    0.17
     ==
    0.17
    <0xC2>
    0.17
    ajima
    0.17
    Act Density 0.477%

    No Known Activations