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

    Neuronpedia

    INDEX
    Explanations

    the word "longtime."

    oai_token-act-pair · gpt-3.5-turbo

    references to duration or permanence of relationships or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     analyse
    -0.84
     analysed
    -0.83
     util
    -0.77
    ised
    -0.76
    opl
    -0.74
    isation
    -0.73
     fulfil
    -0.71
     enqu
    -0.69
     ration
    -0.67
     whilst
    -0.66
    POSITIVE LOGITS
     longtime
    3.27
     longstanding
    1.82
     lifelong
    1.62
     veteran
    1.43
     former
    1.41
     outspoken
    1.18
     perennial
    1.18
     staunch
    1.14
     beloved
    1.12
     embattled
    1.05
    Act Density 0.019%

    No Known Activations