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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a comparison between two options

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that emphasize distinctions or alternatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    awar
    -0.82
    ug
    -0.77
    itiz
    -0.76
    enium
    -0.75
    ns
    -0.72
    hr
    -0.72
    eg
    -0.72
    iatus
    -0.72
    wm
    -0.68
    tan
    -0.68
    POSITIVE LOGITS
     preferably
    0.82
     assuming
    0.70
     apologies
    0.70
     alternatively
    0.70
     optionally
    0.69
     evidenced
    0.68
    Ļ½
    0.66
     perhaps
    0.65
     whichever
    0.65
     allowances
    0.65
    Act Density 0.286%

    No Known Activations