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

    Neuronpedia

    INDEX
    Explanations

    words related to negative reputation or bad behavior

    oai_token-act-pair · gpt-3.5-turbo

    terms related to reputation and pollution

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    kell
    -0.77
    ockets
    -0.68
     joint
    -0.66
    stay
    -0.63
    bees
    -0.63
     heels
    -0.63
    ork
    -0.62
    immer
    -0.62
    appers
    -0.62
    chrom
    -0.59
    POSITIVE LOGITS
    uted
    1.38
    uting
    1.19
    anamo
    1.15
    uality
    0.94
    utes
    0.91
    utations
    0.83
    ously
    0.79
    agate
    0.75
    iating
    0.75
    iously
    0.71
    Act Density 0.009%

    No Known Activations