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

    Neuronpedia

    INDEX
    Explanations

    the word "most" followed by some description

    oai_token-act-pair · gpt-3.5-turbo

    superlative expressions indicating the highest degree of something, particularly "most" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    instead
    -0.82
    roo
    -0.77
    é¾įåĸļ士
    -0.77
    culosis
    -0.75
    ysc
    -0.75
    heid
    -0.75
    æ©
    -0.73
    halla
    -0.73
    uci
    -0.72
    runs
    -0.72
    POSITIVE LOGITS
     obvious
    1.47
     notable
    1.42
     noticeable
    1.38
     important
    1.38
     interesting
    1.33
     salient
    1.32
     intriguing
    1.31
     glaring
    1.29
     worrisome
    1.27
     noteworthy
    1.26
    Act Density 0.048%

    No Known Activations