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

    Neuronpedia

    INDEX
    Explanations

    instances where something is not happening or not being done

    oai_token-act-pair · gpt-3.5-turbo

    negation or denial statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tein
    -0.78
    unks
    -0.69
    aley
    -0.66
    iesel
    -0.66
    hiba
    -0.64
     Line
    -0.64
    ogi
    -0.63
    company
    -0.62
    acker
    -0.62
    ãĤ·ãĥ£
    -0.62
    POSITIVE LOGITS
     necessarily
    1.54
     anymore
    1.02
    etheless
    1.02
    icably
    1.00
    withstanding
    0.97
    icable
    0.96
     enough
    0.96
    eworthy
    0.93
     yet
    0.87
    epad
    0.86
    Act Density 0.062%

    No Known Activations