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

    Neuronpedia

    INDEX
    Explanations

    German words with umlauts, particularly the letter 'ä'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the character "ä"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ORED
    -0.83
     Sussex
    -0.70
     rooting
    -0.65
     logger
    -0.64
     Asians
    -0.63
     Jericho
    -0.62
     Notting
    -0.62
     cavity
    -0.60
     tamp
    -0.59
     rapt
    -0.58
    POSITIVE LOGITS
    ä
    1.46
    inen
    1.32
    ö
    1.01
    ternity
    1.01
    ¢
    0.97
    ë
    0.96
    ·
    0.94
    elsen
    0.92
    ka
    0.91
    ki
    0.90
    Act Density 0.009%

    No Known Activations