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

    Neuronpedia

    INDEX
    Explanations

    the word "well" used as an interjection

    oai_token-act-pair · gpt-3.5-turbo

    expressions of opinion or commentary

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    avascript
    -0.79
    cade
    -0.77
    ebus
    -0.72
    ugal
    -0.72
    ourse
    -0.70
    externalActionCode
    -0.70
    ¶ħ
    -0.69
    luaj
    -0.69
    eryl
    -0.69
    erc
    -0.68
    POSITIVE LOGITS
     dunno
    0.84
     erase
    0.75
     crap
    0.69
     bye
    0.68
     goodbye
    0.67
     tease
    0.65
     Forget
    0.65
     Alternate
    0.64
     boo
    0.64
     yeah
    0.63
    Act Density 0.202%

    No Known Activations