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

    Neuronpedia

    INDEX
    Explanations

    phrases relating to fairness and treatment in a political context

    oai_token-act-pair · gpt-3.5-turbo

    references to expectations and treatment related to political contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     haunt
    -0.68
     Annex
    -0.67
     havoc
    -0.66
     impuls
    -0.65
    mania
    -0.64
     inertia
    -0.61
     unison
    -0.61
    hop
    -0.61
     intrigue
    -0.61
     unprepared
    -0.59
    POSITIVE LOGITS
    taboola
    0.83
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.76
    ãĤ£
    0.75
    ãĤ¦ãĤ¹
    0.72
    ahime
    0.68
    ãĤ§
    0.66
    otally
    0.64
    ãĤ©
    0.64
    ãĥģ
    0.64
    Drug
    0.62
    Act Density 0.049%

    No Known Activations