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

    Neuronpedia

    INDEX
    Explanations

    high-frequency words and conjunctions in discourse

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    è±
    -0.16
    Ĥ¹
    -0.16
    eway
    -0.15
    alog
    -0.15
    EGIN
    -0.15
    fony
    -0.14
    antee
    -0.14
    fout
    -0.14
    ãĤ¹ãĥĨãĤ£
    -0.14
     Wed
    -0.14
    POSITIVE LOGITS
    otts
    0.18
    oux
    0.15
    hausen
    0.14
     Roses
    0.14
    allet
    0.14
    ha
    0.14
     Moore
    0.14
    616
    0.14
     Martins
    0.14
    éħ
    0.14
    Act Density 0.004%

    No Known Activations