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

    Neuronpedia

    INDEX
    Explanations

    strong and impactful action verbs

    oai_token-act-pair · gpt-3.5-turbo

    actions and processes, particularly those implying movement or change

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    avorite
    -0.79
    HOME
    -0.74
     unavoid
    -0.67
     Noise
    -0.65
     impulse
    -0.62
     felon
    -0.62
     Apostle
    -0.60
     spont
    -0.59
     exception
    -0.55
     unfit
    -0.55
    POSITIVE LOGITS
    ed
    1.40
    edIn
    1.39
    ing
    1.35
    uated
    1.14
    arily
    1.10
    ioned
    1.09
    eering
    1.04
    ishly
    1.02
    ized
    1.01
    istered
    1.01
    Act Density 0.383%

    No Known Activations