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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "pretend" or variations of it

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pretend" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     GOODMAN
    -0.67
    cutting
    -0.66
    ighth
    -0.63
    vez
    -0.62
    sbm
    -0.61
    cedented
    -0.61
    stown
    -0.60
    cent
    -0.60
     Citation
    -0.58
    winner
    -0.58
    POSITIVE LOGITS
     innocence
    1.04
     ignorance
    0.78
    inet
    0.70
     allegiance
    0.69
     pas
    0.68
     forgot
    0.68
     insanity
    0.68
     otherwise
    0.64
    idently
    0.64
     inco
    0.63
    Act Density 0.053%

    No Known Activations