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

    Neuronpedia

    INDEX
    Explanations

    references to things being oriented or positioned in an upside-down manner

    oai_token-act-pair · gpt-3.5-turbo

    the concept of "upside" or positive potential in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    naire
    -0.85
    ocally
    -0.78
    nance
    -0.74
    ocal
    -0.74
    ãģ®å®
    -0.73
    ainers
    -0.72
    éļ
    -0.72
    Rush
    -0.71
    ions
    -0.70
    %"
    -0.68
    POSITIVE LOGITS
     upside
    1.18
    ï¸
    0.79
    ward
    0.74
    vable
    0.72
    zzle
    0.71
    verted
    0.70
     infinity
    0.69
     srfAttach
    0.69
     blind
    0.69
    side
    0.67
    Act Density 0.009%

    No Known Activations