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

    Neuronpedia

    INDEX
    Explanations

    phrases related to building, developing, or improving upon something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate building upon previous concepts or foundations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    achy
    -0.91
    ura
    -0.72
    cia
    -0.72
    tu
    -0.66
    opath
    -0.66
     glare
    -0.64
     override
    -0.64
    idential
    -0.63
    asive
    -0.62
    quickShipAvailable
    -0.62
    POSITIVE LOGITS
     existing
    1.15
     previous
    1.07
    existing
    0.96
     predecessor
    0.94
     earlier
    0.91
     predecessors
    0.86
     longstanding
    0.84
     previously
    0.83
     older
    0.80
     tradition
    0.78
    Act Density 0.140%

    No Known Activations