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

    Neuronpedia

    INDEX
    Explanations

    words related to numbers, specifically focusing on multi-digit numbers

    oai_token-act-pair · gpt-3.5-turbo

    references to numerical values, particularly regarding digits and their representations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hire
    -0.89
    nd
    -0.80
    ouf
    -0.77
    Study
    -0.74
    lain
    -0.72
    ModLoader
    -0.71
    nda
    -0.70
    dep
    -0.67
     Shield
    -0.66
    ritz
    -0.65
    POSITIVE LOGITS
    oded
    0.94
    omial
    0.92
     digits
    0.89
    ized
    0.87
    itial
    0.86
    itized
    0.85
    igr
    0.82
    itialized
    0.81
    eteen
    0.81
    ised
    0.81
    Act Density 0.021%

    No Known Activations