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

    Neuronpedia

    INDEX
    Explanations

    terms or phrases that emphasize or draw attention to specific elements, such as through highlighting or emphasizing

    oai_token-act-pair · gpt-3.5-turbo

    the word "highlight" and its variations in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    thood
    -0.72
    cas
    -0.70
    itte
    -0.70
    nery
    -0.69
    entric
    -0.68
    atern
    -0.68
    aah
    -0.67
    onew
    -0.64
    zin
    -0.64
    ja
    -0.64
    POSITIVE LOGITS
     reel
    0.95
     highlights
    0.93
    ibilities
    0.79
    umenthal
    0.78
     highlighting
    0.76
    TextColor
    0.76
    ibility
    0.76
    OWS
    0.74
    iary
    0.74
     warnings
    0.71
    Act Density 0.034%

    No Known Activations