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

    Neuronpedia

    INDEX
    Explanations

    words related to the object "pot," which could be associated with cooking, smoking, or other activities

    oai_token-act-pair · gpt-3.5-turbo

    references to "pot" in terms of marijuana or cooking vessels

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    selves
    -0.73
     Squadron
    -0.71
     Gir
    -0.69
     Vanguard
    -0.69
    aligned
    -0.68
    Force
    -0.68
     Ryder
    -0.67
    âĹ¼
    -0.67
     FN
    -0.67
     RSS
    -0.66
    POSITIVE LOGITS
     pot
    3.73
     pots
    2.69
    pot
    2.19
     Pot
    2.06
    Pot
    1.85
    pots
    1.82
     marijuana
    1.63
     weed
    1.57
    arijuana
    1.51
     Marijuana
    1.38
    Act Density 0.011%

    No Known Activations