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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a preference or an alternative course of action

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "might as well."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    arial
    -0.66
    ostic
    -0.64
    oresc
    -0.63
     Passing
    -0.63
    lining
    -0.60
    ories
    -0.59
    ãģ®å®
    -0.59
    ricting
    -0.58
    doms
    -0.57
    ãĤĮ
    -0.57
    POSITIVE LOGITS
     well
    1.12
    well
    0.90
    ivably
    0.86
    phy
    0.85
    lege
    0.84
     feas
    0.82
    nown
    0.80
     easily
    0.79
     plaus
    0.78
    pired
    0.78
    Act Density 0.049%

    No Known Activations