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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "perfect" and its variants, indicating a focus on excellence or ideal conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Controllo
    -0.59
    dafrika
    -0.56
     wiser
    -0.56
     ancha
    -0.54
     efectivo
    -0.52
     gruesa
    -0.50
     NUKAT
    -0.50
     fortement
    -0.50
     igjen
    -0.49
    tubers
    -0.49
    POSITIVE LOGITS
     Perfect
    1.09
     perfect
    1.08
    Perfect
    1.07
     PERFECT
    1.06
     imperfect
    1.03
     imperfection
    1.02
    perfect
    1.01
    PERFECT
    1.00
     imperfections
    0.93
     IDEAL
    0.92
    Act Density 0.046%

    No Known Activations