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

    Neuronpedia

    INDEX
    Explanations

    words related to completing actions or obligations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "fulfill" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     gradient
    -0.67
    Heat
    -0.66
    hill
    -0.65
    heat
    -0.65
    face
    -0.63
    bug
    -0.63
    rr
    -0.63
    Tip
    -0.63
    Lou
    -0.63
     Barber
    -0.62
    POSITIVE LOGITS
     fulfilled
    3.35
     fulfil
    3.00
     fulfill
    2.94
     fulfillment
    2.93
     fulfilling
    2.56
     fulf
    2.06
     satisfy
    1.37
     satisf
    1.17
     satisfied
    1.12
     satisfying
    1.07
    Act Density 0.018%

    No Known Activations