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

    Neuronpedia

    INDEX
    Explanations

    titles or roles related to seniority or leadership positions

    oai_token-act-pair · gpt-3.5-turbo

    titles and positions held by individuals, particularly in academic and professional contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     peril
    -0.96
     reper
    -0.90
     queens
    -0.89
     compromise
    -0.87
     dwarves
    -0.87
     queen
    -0.85
     trap
    -0.85
     palace
    -0.85
     plaus
    -0.85
     doom
    -0.82
    POSITIVE LOGITS
     Manager
    1.63
     Coordinator
    1.57
     Director
    1.54
     Editor
    1.52
     Partner
    1.50
     Advisor
    1.47
     Producer
    1.46
     Owner
    1.46
     Analyst
    1.45
     Adviser
    1.45
    Act Density 0.182%

    No Known Activations