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

    Neuronpedia

    INDEX
    Explanations

    instances where something is singled out as unique or exclusive

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating exclusivity or uniqueness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tnc
    -0.65
    ahime
    -0.65
    èĪ
    -0.61
     whirlwind
    -0.59
    enture
    -0.59
     [|
    -0.57
    description
    -0.57
     masse
    -0.56
     DRAG
    -0.56
    staking
    -0.56
    POSITIVE LOGITS
     remaining
    1.12
     ever
    1.12
     EVER
    1.08
     viable
    1.07
     imaginable
    1.05
     survives
    1.04
     conceivable
    1.03
     usable
    1.02
    ever
    1.02
     anymore
    1.02
    Act Density 0.276%

    No Known Activations