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

    Neuronpedia

    INDEX
    Explanations

    pronouns and questions related to personal beliefs or opinions

    oai_token-act-pair · gpt-3.5-turbo

    pronouns and references to personal identity or relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    opens
    -0.79
    accompan
    -0.75
    answered
    -0.70
    ufact
    -0.69
    ī
    -0.64
    ansas
    -0.63
    itiz
    -0.62
    handled
    -0.61
    aby
    -0.61
    PDATE
    -0.61
    POSITIVE LOGITS
     owe
    1.23
     deserve
    1.20
     intend
    1.12
     belong
    1.10
     need
    1.10
     mean
    1.10
     want
    1.06
     qualify
    1.02
     realise
    1.01
     know
    1.00
    Act Density 0.103%

    No Known Activations