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

    Neuronpedia

    INDEX
    Explanations

    superlatives indicating the most extreme degree or ranking of something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "most" used to emphasize extremes or significance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    æ©
    -0.74
    pload
    -0.70
    heid
    -0.67
    oak
    -0.66
    rompt
    -0.66
     Films
    -0.65
    olor
    -0.64
    halla
    -0.63
    instead
    -0.63
     Travels
    -0.63
    POSITIVE LOGITS
     important
    1.02
    likely
    0.93
     obvious
    0.92
     profitable
    0.91
     notable
    0.90
     interesting
    0.88
     prominent
    0.88
     likely
    0.88
     plausible
    0.88
     exciting
    0.87
    Act Density 0.063%

    No Known Activations