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

    Neuronpedia

    INDEX
    Explanations

    irregular punctuation and formatting elements in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    irit
    -0.16
    eca
    -0.15
    eyer
    -0.15
    rie
    -0.14
    abel
    -0.14
    aper
    -0.14
    bum
    -0.14
    ease
    -0.14
    ús
    -0.14
    lass
    -0.13
    POSITIVE LOGITS
     Brook
    0.16
    phans
    0.15
    ungi
    0.15
     bro
    0.15
     entirety
    0.14
    adÃŃ
    0.14
    丸
    0.14
    ãĤ«ãĥ«
    0.13
    éré
    0.13
    ilik
    0.13
    Act Density 0.047%

    No Known Activations