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

    Neuronpedia

    INDEX
    Explanations

    mentions of specific items or products, such as "BMW auto parts" or "knit scarves."

    oai_token-act-pair · gpt-3.5-turbo

    references to various specific items and concepts, including brands, operations, and attributes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     gracious
    -0.70
     captive
    -0.66
     Gw
    -0.66
     disbelief
    -0.65
     intrinsic
    -0.65
     hindsight
    -0.64
     drunk
    -0.63
     interchange
    -0.63
     Suc
    -0.63
     inev
    -0.62
    POSITIVE LOGITS
    atures
    1.23
    ensions
    1.14
    ules
    1.14
    aters
    1.12
    elines
    1.11
    ences
    1.10
    rations
    1.07
    vals
    1.06
    estones
    1.05
    istries
    1.05
    Act Density 0.278%

    No Known Activations