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

    Neuronpedia

    INDEX
    Explanations

    keywords related to conventional or traditional methods, structures, or concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to conventional standards or practices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hop
    -0.75
    haw
    -0.71
    oning
    -0.70
    ander
    -0.68
    hov
    -0.66
    oran
    -0.66
    hoff
    -0.66
    hner
    -0.65
    atoon
    -0.65
    gur
    -0.64
    POSITIVE LOGITS
    ventional
    1.04
     wisdom
    0.95
     conventional
    0.94
    etheless
    0.87
    itionally
    0.85
    ually
    0.80
    ijah
    0.78
    ization
    0.78
    ensical
    0.77
    aneously
    0.76
    Act Density 0.009%

    No Known Activations