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

    Neuronpedia

    INDEX
    Explanations

    references to pixels, their characteristics, and operations related to pixels

    oai_token-act-pair · gpt-3.5-turbo

    references to pixels in graphical content

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    shore
    -0.77
    avanaugh
    -0.75
    redo
    -0.73
    doms
    -0.70
    urst
    -0.68
    UD
    -0.67
    etheus
    -0.66
    spell
    -0.66
    dule
    -0.65
    igham
    -0.65
    POSITIVE LOGITS
     pixel
    0.94
    ixels
    0.92
     pixels
    0.90
     shader
    0.81
    Pixel
    0.81
     shading
    0.74
    agraph
    0.74
    apixel
    0.72
     OLED
    0.70
     fonts
    0.70
    Act Density 0.009%

    No Known Activations