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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "draw" and its variants

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    glie
    -0.54
     accett
    -0.53
    keit
    -0.52
    nativeElement
    -0.50
     accepté
    -0.50
     Prestige
    -0.49
     ambientales
    -0.49
    lepší
    -0.49
    Amin
    -0.48
     /\.(
    -0.48
    POSITIVE LOGITS
     attention
    0.97
     conclusions
    0.88
     parallels
    0.87
    attention
    0.85
     drawn
    0.78
     Draws
    0.78
    aarrggbb
    0.77
     Drawn
    0.76
     straws
    0.76
     draws
    0.75
    Act Density 0.074%

    No Known Activations