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

    Neuronpedia

    INDEX
    Explanations

    references to specific shapes, particularly rectangles

    oai_token-act-pair · gpt-3.5-turbo

    references to specific shapes, objects, or categories of items related to clothing and meals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bies
    -0.83
    hab
    -0.80
    esty
    -0.78
    axter
    -0.77
    hus
    -0.73
    dm
    -0.67
    itability
    -0.65
    quarters
    -0.65
    govern
    -0.65
    lins
    -0.64
    POSITIVE LOGITS
     Kinnikuman
    0.79
    naire
    0.78
    LOAD
    0.74
     è£ıç
    0.70
    OPLE
    0.66
    geist
    0.64
    ographed
    0.64
     Cerberus
    0.63
    é¾įåĸļ士
    0.63
     Gladiator
    0.63
    Act Density 0.035%

    No Known Activations