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

    Neuronpedia

    INDEX
    Explanations

    phrases related to moral contrasts, especially focusing on the concepts of good, bad, right, and wrong

    oai_token-act-pair · gpt-3.5-turbo

    phrases and concepts related to moral and ethical dilemmas, particularly contrasting good and bad

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mud
    -0.75
    76561
    -0.73
    rh
    -0.70
    culosis
    -0.67
    thur
    -0.67
    ãĥĹ
    -0.67
    forts
    -0.66
    atl
    -0.65
    gow
    -0.65
    vice
    -0.64
    POSITIVE LOGITS
     depending
    1.26
     alike
    1.12
    depending
    1.02
     respectively
    0.90
     sides
    0.88
     eras
    0.81
     halves
    0.78
     dich
    0.77
     perspectives
    0.77
     administrations
    0.77
    Act Density 0.265%

    No Known Activations