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

    Neuronpedia

    INDEX
    Explanations

    phrases referring to groups of people or individuals

    oai_token-act-pair · gpt-3.5-turbo

    repetitive phrases or structures related to people

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Inventory
    -0.79
    ¨
    -0.75
     Goodbye
    -0.71
    RAY
    -0.68
     Solution
    -0.67
     Ensure
    -0.66
     Deter
    -0.65
     Grape
    -0.64
     Farn
    -0.63
     Peoples
    -0.63
    POSITIVE LOGITS
     fortunate
    1.00
     interested
    0.96
     lucky
    0.96
     unlucky
    0.93
     willing
    0.91
     genuinely
    0.91
     addicted
    0.91
     offended
    0.90
     accustomed
    0.89
     intimately
    0.89
    Act Density 0.151%

    No Known Activations