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

    Neuronpedia

    INDEX
    Explanations

    phrases related to honoring or being privileged

    oai_token-act-pair · gpt-3.5-turbo

    terms related to recognition and positive sentiments about achievements or good fortune

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     cheat
    -0.78
    idine
    -0.72
    band
    -0.72
    bang
    -0.71
    bender
    -0.71
    hang
    -0.70
    valid
    -0.69
    stress
    -0.68
    ster
    -0.68
     ballistic
    -0.67
    POSITIVE LOGITS
    quished
    0.75
    dinand
    0.73
     Lauder
    0.72
     Seym
    0.70
    Reviewer
    0.69
    REAM
    0.69
     fortunate
    0.69
     upbringing
    0.68
    herty
    0.68
     privileged
    0.68
    Act Density 0.026%

    No Known Activations