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

    Neuronpedia

    INDEX
    Explanations

    phrases related to hierarchy or sequence

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate hierarchy or order within a system

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iland
    -0.71
     contrace
    -0.68
    >[
    -0.66
     Sachs
    -0.63
    zinski
    -0.63
    furt
    -0.61
    internet
    -0.61
    gow
    -0.59
     prevailing
    -0.59
     ',
    -0.57
    POSITIVE LOGITS
     magnitude
    0.97
     Cthulhu
    0.96
     precedence
    0.96
    lege
    0.88
     Duty
    0.87
    rontal
    0.86
     duty
    0.81
    aiden
    0.79
    redo
    0.74
    alions
    0.74
    Act Density 0.097%

    No Known Activations