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

    Neuronpedia

    INDEX
    Explanations

    proper names or entities, particularly when accompanied by a possessive apostrophe 's

    oai_token-act-pair · gpt-3.5-turbo

    certain characters or symbols used in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     democracy
    -0.67
     ``
    -0.62
     Mobil
    -0.60
     dignity
    -0.58
     division
    -0.58
     uranium
    -0.57
     Democracy
    -0.57
     neut
    -0.56
     chart
    -0.56
     ................
    -0.55
    POSITIVE LOGITS
    s
    1.48
    t
    1.06
    sn
    1.01
    dq
    1.01
    shall
    0.99
    sat
    0.97
    d
    0.97
    sed
    0.96
    should
    0.95
    Pg
    0.95
    Act Density 0.302%

    No Known Activations