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

    Neuronpedia

    INDEX
    Explanations

    Twitter handles and online usernames

    oai_token-act-pair · gpt-3.5-turbo

    Twitter handles and references to specific names or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Þ
    -0.99
    ñ
    -0.92
    ò
    -0.90
    ß
    -0.83
     pione
    -0.83
     oun
    -0.82
     eleph
    -0.80
     conflic
    -0.79
     subur
    -0.79
     awa
    -0.78
    POSITIVE LOGITS
    t
    1.16
    tty
    1.04
    ts
    1.03
    tt
    0.98
    td
    0.92
    tis
    0.91
    dt
    0.91
    plet
    0.91
    tp
    0.91
    ton
    0.90
    Act Density 0.211%

    No Known Activations