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

    Neuronpedia

    INDEX
    Explanations

    the adjective "moderate" and related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to moderation in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Lust
    -0.72
    STON
    -0.70
    hon
    -0.70
    borne
    -0.69
    arium
    -0.69
    stone
    -0.69
    yang
    -0.69
    RIP
    -0.68
    HCR
    -0.68
    GV
    -0.68
    POSITIVE LOGITS
    erate
    1.10
    yip
    0.93
     minded
    0.92
     moderate
    0.91
    moderate
    0.91
     leaning
    0.80
    mble
    0.78
     medi
    0.77
     sized
    0.77
    »Ĵ
    0.77
    Act Density 0.012%

    No Known Activations