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

    Neuronpedia

    INDEX
    Explanations

    numerical strings in a specific format

    oai_token-act-pair · gpt-3.5-turbo

    numerical values or identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    efully
    -0.80
     millenn
    -0.74
     veter
    -0.71
    anooga
    -0.71
     tremend
    -0.70
    querque
    -0.69
    istry
    -0.67
    ionics
    -0.66
    inery
    -0.66
     thous
    -0.65
    POSITIVE LOGITS
    GMT
    1.13
    .............
    1.03
     GMT
    0.96
    bps
    0.93
    608
    0.87
    265
    0.85
    341
    0.85
    408
    0.85
    1001
    0.85
    704
    0.85
    Act Density 0.057%

    No Known Activations