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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating regret or hindsight

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "should have."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     hostage
    -0.77
     maze
    -0.68
     fireball
    -0.68
     craving
    -0.64
    Mem
    -0.64
    catentry
    -0.60
     hots
    -0.59
     Cold
    -0.59
     ceiling
    -0.58
     muse
    -0.58
    POSITIVE LOGITS
     gotten
    0.97
     been
    0.95
    been
    0.87
    gotten
    0.86
     acted
    0.86
     behaved
    0.83
     done
    0.79
     Been
    0.77
    ĭ
    0.75
     taken
    0.75
    Act Density 0.057%

    No Known Activations