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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the best or safest option in a given scenario

    oai_token-act-pair · gpt-3.5-turbo

    phrases that suggest the best or optimal choices or solutions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    lov
    -0.70
    abuse
    -0.68
    cart
    -0.67
    Kit
    -0.66
    clerosis
    -0.66
    pine
    -0.65
    older
    -0.64
    mat
    -0.61
    ipel
    -0.61
     LOG
    -0.61
    POSITIVE LOGITS
     way
    1.35
     solution
    1.33
     option
    1.21
     alternative
    1.12
     antidote
    1.11
     answer
    1.10
     thing
    1.08
     recourse
    1.06
     alternatives
    1.05
     ways
    1.04
    Act Density 0.196%

    No Known Activations