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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating contradiction or opposite ideas

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "contrary" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    bers
    -0.74
    utters
    -0.72
    asp
    -0.69
    hens
    -0.66
    mberg
    -0.65
    artney
    -0.63
    ETF
    -0.62
    anon
    -0.62
    ardi
    -0.62
    ARD
    -0.62
    POSITIVE LOGITS
    etheless
    1.00
     contrary
    0.92
     guiActiveUn
    0.85
     undermin
    0.83
     minded
    0.80
     materially
    0.77
     Contrary
    0.75
    lihood
    0.73
     substant
    0.72
     notwithstanding
    0.71
    Act Density 0.010%

    No Known Activations