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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences where emphasis is added

    oai_token-act-pair · gpt-3.5-turbo

    terms related to the emphasis placed on specific points or definitions in a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    heny
    -0.69
    vati
    -0.68
    duc
    -0.67
    jri
    -0.67
    roups
    -0.65
    existence
    -0.64
    kus
    -0.63
    '>
    -0.63
    ttle
    -0.63
     Skydragon
    -0.62
    POSITIVE LOGITS
     mine
    1.10
     omitted
    1.06
     ours
    1.04
     ital
    0.97
     supplied
    0.95
     redacted
    0.93
     emphasis
    0.93
     added
    0.92
     hers
    0.91
     theirs
    0.89
    Act Density 0.064%

    No Known Activations