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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating causality

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "due" indicating causation or reason

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arest
    -0.82
    =-=-=-=-=-=-=-=-
    -0.78
    asp
    -0.75
    arro
    -0.74
    hov
    -0.73
     Sheep
    -0.71
    anon
    -0.71
     Petra
    -0.68
    agate
    -0.67
    oola
    -0.65
    POSITIVE LOGITS
     diligence
    1.17
    giving
    0.86
    gers
    0.81
     cancell
    0.81
     releg
    0.75
    itiz
    0.73
    */(
    0.71
     dilig
    0.70
     deadlines
    0.69
     periods
    0.68
    Act Density 0.021%

    No Known Activations