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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the word "aud" with different numerical values

    oai_token-act-pair · gpt-3.5-turbo

    references to sound or audio

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ragnarok
    -0.79
     bis
    -0.74
     Copenhagen
    -0.69
     Helsinki
    -0.68
     Warriors
    -0.68
     MET
    -0.66
     eternal
    -0.65
     Plenty
    -0.64
     Stockholm
    -0.64
     Loki
    -0.63
    POSITIVE LOGITS
    aud
    1.17
    issance
    1.04
    ience
    1.04
    iences
    1.02
    eatures
    1.00
    ible
    0.95
    itism
    0.95
    iance
    0.95
    escription
    0.94
    iosyn
    0.90
    Act Density 0.004%

    No Known Activations