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

    Neuronpedia

    INDEX
    Explanations

    descriptive adjectives or phrases indicating evaluation or opinion

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "it"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     paran
    -0.60
    evil
    -0.59
     Dayton
    -0.56
     Eighth
    -0.55
    anton
    -0.54
     heartbeat
    -0.52
    kie
    -0.52
    ————————
    -0.51
    former
    -0.51
    VA
    -0.51
    POSITIVE LOGITS
     beh
    1.27
     seems
    1.26
     begs
    1.11
     appears
    1.09
     becomes
    1.06
     seemed
    1.04
    zik
    1.03
    chy
    1.01
     Seems
    1.00
    unes
    0.99
    Act Density 0.193%

    No Known Activations