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

    Neuronpedia

    INDEX
    Explanations

    phrases describing intensity or emphasis

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating strong emphasis or intensity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Mechdragon
    -0.67
     Cosponsors
    -0.64
     Compliance
    -0.64
     Credit
    -0.62
    soDeliveryDate
    -0.61
     Cipher
    -0.61
    inventoryQuantity
    -0.61
    luaj
    -0.61
     Landing
    -0.61
     Deng
    -0.61
    POSITIVE LOGITS
     uh
    0.89
     um
    0.85
    nery
    0.77
    ivably
    0.72
    ribly
    0.71
    itely
    0.70
     positively
    0.69
    really
    0.68
     VERY
    0.68
     ah
    0.68
    Act Density 0.059%

    No Known Activations