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

    Neuronpedia

    INDEX
    Explanations

    say "math" operators

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    )-
    1.39
    ]-
    1.32
    |-
    1.31
    }-
    1.27
     )-
    1.22
    \}-
    1.19
    >-
    1.17
    ())+
    1.11
    })-
    1.08
    }-(
    1.07
    POSITIVE LOGITS
    +
    0.76
     +
    0.74
     soup
    0.69
     Soup
    0.66
    toi
    0.65
    躾
    0.64
    apun
    0.64
     electrolytes
    0.64
     intoxication
    0.62
     архів
    0.62
    Act Density 0.341%

    No Known Activations