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

    Neuronpedia

    INDEX
    Explanations

    Keywords related to expertise or proficiency

    oai_token-act-pair · gpt-3.5-turbo

    terms that denote expertise and skill

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    orate
    -0.71
    ndum
    -0.69
     Skydragon
    -0.65
    idium
    -0.64
    ãĤ¡
    -0.64
    Neigh
    -0.63
    izons
    -0.61
    hea
    -0.61
     Territories
    -0.60
    idia
    -0.59
    POSITIVE LOGITS
     negotiator
    1.13
     manip
    1.09
     tact
    1.08
     liar
    1.02
     crafts
    0.93
     swords
    0.91
     lingu
    0.91
     pian
    0.91
     strategist
    0.89
     improvis
    0.89
    Act Density 0.278%

    No Known Activations