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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "specifically" in the text

    oai_token-act-pair · gpt-3.5-turbo

    details and clarifications within a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Tycoon
    -0.82
    =-=-=-=-=-=-=-=-
    -0.72
     Utt
    -0.72
    anon
    -0.70
    tein
    -0.69
     Oaks
    -0.69
     Isles
    -0.69
    Afee
    -0.68
    ulton
    -0.67
    Sad
    -0.67
    POSITIVE LOGITS
     tailored
    1.11
    cedented
    0.93
     exempted
    0.93
     targeted
    0.90
     specifically
    0.89
     suited
    0.88
     identifiable
    0.88
     appropriated
    0.85
     appropri
    0.82
     pinpoint
    0.82
    Act Density 0.005%

    No Known Activations