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

    Neuronpedia

    INDEX
    Explanations

    keywords related to ensuring proper handling or management of things

    oai_token-act-pair · gpt-3.5-turbo

    the term "properly" in various contexts indicating appropriate actions or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ista
    -0.82
    iens
    -0.80
    ————————
    -0.75
    =-=-=-=-=-=-=-=-
    -0.74
    oleon
    -0.74
    raltar
    -0.68
    ocene
    -0.68
    İ
    -0.66
    hyde
    -0.66
    ists
    -0.65
    POSITIVE LOGITS
     behaved
    0.97
     positioned
    0.92
     formatted
    0.89
     equipped
    0.87
     handled
    0.86
     clothed
    0.86
     aligned
    0.86
     cared
    0.84
     suited
    0.84
     dispose
    0.83
    Act Density 0.019%

    No Known Activations