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

    Neuronpedia

    INDEX
    Explanations

    comparison operators and expressions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     equal
    1.29
    equal
    1.08
     equally
    1.08
     Equal
    0.99
    Equal
    0.95
     equals
    0.95
     gleich
    0.87
     равно
    0.83
     igual
    0.82
    equals
    0.82
    POSITIVE LOGITS
     >
    1.56
     (>
    1.51
     exceeds
    1.44
    ＞
    1.40
     $>
    1.37
    $>
    1.37
     exceed
    1.34
    1.30
     surpasses
    1.29
    >
    1.28
    Act Density 0.300%

    No Known Activations