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

    Neuronpedia

    INDEX
    Explanations

    phrases containing the word "Keep" used as a command or suggestion

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases indicating continuity or reminders to stay engaged

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     accomplished
    -0.74
     Gall
    -0.66
     debut
    -0.65
     equ
    -0.64
     spectrum
    -0.64
     revol
    -0.63
     experienced
    -0.63
     obese
    -0.63
     tang
    -0.62
     about
    -0.62
    POSITIVE LOGITS
    Keep
    3.54
    keep
    1.62
    Stay
    1.59
    Get
    1.58
    skip
    1.51
    Keeping
    1.48
    Newsletter
    1.43
    Always
    1.33
    Continue
    1.31
    Put
    1.31
    Act Density 0.022%

    No Known Activations