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

    Neuronpedia

    INDEX
    Explanations

    phrases related to high quality or top-tier characteristics

    oai_token-act-pair · gpt-3.5-turbo

    references to global or world-class achievements and concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ell
    -0.70
     Mehran
    -0.69
     Removal
    -0.67
    essee
    -0.67
    ï¸ı
    -0.64
     interval
    -0.64
     child
    -0.64
     HI
    -0.62
     Toro
    -0.62
     ABE
    -0.61
    POSITIVE LOGITS
    wide
    1.42
    Wide
    1.10
    famous
    1.09
    saving
    1.03
    leading
    0.99
    level
    0.96
    readable
    0.96
    oriented
    0.95
    class
    0.92
    neutral
    0.92
    Act Density 0.062%

    No Known Activations