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

    Neuronpedia

    INDEX
    Explanations

    references to box and data structure terminology in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    withstanding
    -0.98
     pleaſure
    -0.97
    ModelAdmin
    -0.92
     Majefty
    -0.86
     purpoſe
    -0.80
     secondaires
    -0.79
     Nacionales
    -0.78
     scuro
    -0.78
     ſtill
    -0.78
    LEIA
    -0.77
    POSITIVE LOGITS
     box
    2.52
     Box
    2.45
     BOX
    2.37
    box
    2.31
     boxes
    2.31
    BOX
    2.18
    Box
    2.16
     Boxes
    2.08
    boxes
    2.01
    Boxes
    1.76
    Act Density 0.101%

    No Known Activations