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

    Neuronpedia

    INDEX
    Explanations

    words or phrases expressing contrast or contradiction

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating contrast or concession

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    asus
    -0.70
     burg
    -0.65
    icles
    -0.65
    otaur
    -0.64
    mer
    -0.63
    med
    -0.62
     helmets
    -0.60
     GEAR
    -0.58
     Cathedral
    -0.58
    nec
    -0.58
    POSITIVE LOGITS
    theless
    0.92
    chart
    0.82
     nonetheless
    0.79
    NESS
    0.78
     survives
    0.75
     nevertheless
    0.73
     conclud
    0.73
     prevail
    0.73
     persisted
    0.72
     notwithstanding
    0.72
    Act Density 0.008%

    No Known Activations