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

    Neuronpedia

    INDEX
    Explanations

    countries or entities responsible for actions or events

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases that indicate responsibility or blame in a context of conflict or actions taken

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     RELEASE
    -0.72
    lif
    -0.70
    adesh
    -0.68
    udo
    -0.66
    ione
    -0.64
     Lif
    -0.64
     Habit
    -0.64
    Ut
    -0.62
     Congratulations
    -0.62
    UNE
    -0.62
    POSITIVE LOGITS
     responsible
    1.18
     blame
    1.08
     culprit
    1.07
     contributing
    1.05
    responsible
    1.00
     culp
    0.99
     influencing
    0.98
     contribut
    0.96
     perpetrators
    0.92
     motivating
    0.92
    Act Density 0.972%

    No Known Activations