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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating the presence of potential or possibility

    oai_token-act-pair · gpt-3.5-turbo

    references to potential, particularly in contexts suggesting capabilities or possibilities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tein
    -0.74
     Payton
    -0.72
    ĪĴ
    -0.71
    aten
    -0.68
    OTO
    -0.68
    phia
    -0.68
    cise
    -0.68
     Engel
    -0.67
    cipline
    -0.67
    cloth
    -0.66
    POSITIVE LOGITS
    ities
    1.07
     pitfalls
    0.95
    ity
    0.88
     adversaries
    0.85
     usefulness
    0.84
    externalActionCode
    0.82
    ibilities
    0.78
     hazards
    0.78
    atility
    0.78
     payoff
    0.77
    Act Density 0.045%

    No Known Activations