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

    Neuronpedia

    INDEX
    Explanations

    the word "cigarette" at different points in the text

    oai_token-act-pair · gpt-3.5-turbo

    references to cigarettes and smoking

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    */(
    -0.85
    pmwiki
    -0.76
    isite
    -0.74
    ede
    -0.72
    alon
    -0.71
    herty
    -0.67
    UFC
    -0.66
    eele
    -0.66
     Yates
    -0.66
    ology
    -0.65
    POSITIVE LOGITS
    arette
    1.30
    arettes
    1.26
     cigarettes
    1.22
     cigarette
    1.12
     cig
    1.11
     smoking
    1.11
     smokers
    1.10
     smoker
    1.09
     smoked
    0.97
     puff
    0.96
    Act Density 0.009%

    No Known Activations