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

    Neuronpedia

    INDEX
    Explanations

    transitions signaling a shift in focus or a counterpoint in an argument

    oai_token-act-pair · gpt-3.5-turbo

    the word "nevertheless" and its variations, indicating contrasts or exceptions in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     burg
    -0.76
    bons
    -0.71
    icles
    -0.66
    amps
    -0.64
    mer
    -0.62
    raz
    -0.61
    isation
    -0.60
    med
    -0.60
    bind
    -0.59
     Cathedral
    -0.58
    POSITIVE LOGITS
    theless
    1.06
    NESS
    0.77
     rehe
    0.76
     notwithstanding
    0.74
     conclud
    0.74
    entimes
    0.74
     guiActiveUn
    0.73
     survives
    0.72
    hovah
    0.71
    xual
    0.71
    Act Density 0.005%

    No Known Activations