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

    Neuronpedia

    INDEX
    Explanations

    references to boxes and their attributes in defined structures or layouts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    withstanding
    -0.90
     pleaſure
    -0.86
    aarrggbb
    -0.85
    PerformLayout
    -0.85
    ModelAdmin
    -0.84
    DSA
    -0.81
     Ciri
    -0.75
     scuro
    -0.74
    */),
    -0.73
    rslt
    -0.73
    POSITIVE LOGITS
     BOX
    1.99
     Box
    1.97
     box
    1.95
     boxes
    1.90
    BOX
    1.79
    box
    1.78
     Boxes
    1.73
    Box
    1.70
    boxes
    1.61
    Boxes
    1.47
    Act Density 0.048%

    No Known Activations