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

    Neuronpedia

    INDEX
    Explanations

    opinions or evaluations about specific topics or individuals

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    Es
    -1.10
    Els
    -1.04
    ¯¯¯¯
    -1.02
    Ñı
    -0.91
    burgh
    -0.87
    Balt
    -0.87
    aqu
    -0.87
    Guest
    -0.85
    Animal
    -0.85
    berry
    -0.85
    POSITIVE LOGITS
     Hilbert
    1.05
    »Ĵ
    0.94
    rha
    0.92
     disag
    0.91
     hindsight
    0.90
     hypot
    0.88
     retrospect
    0.87
     graph
    0.85
     numer
    0.83
     carefully
    0.83
    Act Density 0.873%

    No Known Activations