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

    Neuronpedia

    INDEX
    Explanations

    instances of irony and humorous contradictions in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    overe
    -0.16
     Hizmetleri
    -0.14
     spis
    -0.14
    ãĥĪãĥª
    -0.14
    ramer
    -0.14
     hopefully
    -0.14
    ream
    -0.14
    .charCodeAt
    -0.13
    Hopefully
    -0.13
     kennenlernen
    -0.13
    POSITIVE LOGITS
     exactly
    0.24
     actually
    0.22
     precisely
    0.20
     именно
    0.19
    actually
    0.18
     neither
    0.18
     none
    0.17
    ç«Ł
    0.17
    wort
    0.17
     also
    0.17
    Act Density 0.135%

    No Known Activations