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

    Neuronpedia

    INDEX
    Explanations

    names or partial names containing "K" followed by a high-activation-value vowel like 'a', 'o', or 'u'

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns, specifically names

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    IBLE
    -0.80
    ãĥ¼ãĥĨãĤ£
    -0.69
    Ö¼
    -0.68
    天
    -0.66
    ional
    -0.64
    eric
    -0.62
     à¨
    -0.62
    UCT
    -0.60
    د
    -0.59
    س
    -0.58
    POSITIVE LOGITS
    inski
    0.98
    lyak
    0.94
    lein
    0.90
    lov
    0.90
    lar
    0.90
    insky
    0.88
    laus
    0.86
    patrick
    0.86
    istani
    0.83
    anski
    0.83
    Act Density 0.077%

    No Known Activations