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

    Neuronpedia

    INDEX
    Explanations

    phrases or keywords related to expressing preferences

    oai_token-act-pair · gpt-3.5-turbo

    expressions of personal preference

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    breakers
    -0.78
    pack
    -0.74
    brance
    -0.74
    breaks
    -0.74
    effects
    -0.73
    gren
    -0.72
    angers
    -0.68
    breaker
    -0.67
    meta
    -0.65
    bleacher
    -0.64
    POSITIVE LOGITS
    itism
    0.75
    rals
    0.74
     Mistress
    0.65
    quickShipAvailable
    0.65
    yip
    0.62
     ratios
    0.61
    ancy
    0.61
     embodiment
    0.61
     embodiments
    0.60
     preferring
    0.60
    Act Density 0.018%

    No Known Activations