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

    Neuronpedia

    INDEX
    Explanations

    words related to being an exceptional case or not following a general rule

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to exceptions to rules or generalizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     destro
    -0.72
    DCS
    -0.71
    espie
    -0.61
     goodbye
    -0.60
     juven
    -0.60
    Progress
    -0.58
     tyr
    -0.58
    ching
    -0.57
    yang
    -0.57
    ched
    -0.56
    POSITIVE LOGITS
    arily
    0.89
    ĸļ
    0.88
    ality
    0.87
    ional
    0.84
    als
    0.80
     exceptions
    0.80
    abl
    0.79
    aux
    0.75
     exception
    0.72
    izzle
    0.71
    Act Density 0.051%

    No Known Activations