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

    Neuronpedia

    INDEX
    Explanations

    verb "to be" (am, is, are, was, were) followed by a present participle (base form of the verb + 'ing')

    oai_token-act-pair · gpt-3.5-turbo

    contractions and forms of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    =-=-
    -0.85
    olit
    -0.67
    ggles
    -0.63
    iencies
    -0.61
    intestinal
    -0.60
    loe
    -0.60
    rosso
    -0.59
    姫
    -0.59
    cture
    -0.59
    rontal
    -0.58
    POSITIVE LOGITS
     doing
    1.31
     proposing
    1.14
     saying
    1.02
     planning
    1.01
     aiming
    1.00
     trying
    1.00
     experiencing
    0.99
     thinking
    0.98
     capable
    0.98
     Doing
    0.97
    Act Density 0.065%

    No Known Activations