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

    Neuronpedia

    INDEX
    Explanations

    numerical or coded data

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Numbers (often "7") after symbols

    np_acts-logits-general · gemini-2.0-flash

    numbers and 7

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Borne
    -0.52
    nial
    -0.51
    NOON
    -0.47
    urum
    -0.47
     Petitioner
    -0.46
    공
    -0.46
    ilers
    -0.46
     Roh
    -0.45
    titut
    -0.45
     □
    -0.44
    POSITIVE LOGITS
    7
    1.34
     seven
    1.07
     seventh
    1.04
     Seven
    1.03
     Seventh
    0.95
     tujuh
    0.93
     VII
    0.92
    Seven
    0.89
    seven
    0.88
     zeven
    0.86
    Act Density 0.625%

    No Known Activations