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

    Neuronpedia

    INDEX
    Explanations

    positive adjectives denoting quality or desirability

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing good quality or positive attributes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    doms
    -0.88
    letters
    -0.81
    ravings
    -0.80
    onduct
    -0.79
    bots
    -0.78
    orders
    -0.73
    ancies
    -0.73
    venants
    -0.73
    Muslims
    -0.73
     Plans
    -0.73
    POSITIVE LOGITS
     reminder
    1.35
     addition
    1.27
     distraction
    1.15
     example
    1.14
     tool
    1.13
     choice
    1.08
     antidote
    1.06
     opportunity
    1.05
     way
    1.05
     indicator
    1.04
    Act Density 0.135%

    No Known Activations