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

    Neuronpedia

    INDEX
    Explanations

    phrases related to exceeding or surpassing something in performance or comparison

    oai_token-act-pair · gpt-3.5-turbo

    terms that indicate exceeding or outperforming

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Refuge
    -0.68
     Origin
    -0.66
    Accessory
    -0.65
     ACTION
    -0.65
     Horror
    -0.65
     Geneva
    -0.65
    Interstitial
    -0.65
    Downloadha
    -0.62
     Goods
    -0.61
     Geo
    -0.61
    POSITIVE LOGITS
    mented
    1.10
    uple
    0.90
    ighed
    0.86
    ceed
    0.85
    İĭ
    0.83
    Ķ
    0.83
    lex
    0.82
    rolet
    0.82
    pling
    0.79
    pload
    0.78
    Act Density 0.015%

    No Known Activations