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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating representation or action taken on someone's behalf

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating representation or advocacy for others

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ined
    -0.60
    itals
    -0.58
     Sapp
    -0.58
     crotch
    -0.58
     diapers
    -0.58
    bern
    -0.57
     Conway
    -0.57
     plum
    -0.57
     stir
    -0.56
     Marathon
    -0.56
    POSITIVE LOGITS
    selves
    0.81
     behalf
    0.76
    avement
    0.75
    anding
    0.72
    giving
    0.72
    acion
    0.69
    ours
    0.68
    oux
    0.68
    uary
    0.65
    edIn
    0.65
    Act Density 0.023%

    No Known Activations