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

    Neuronpedia

    INDEX
    Explanations

    strings representing image sources from photography services

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     gir
    -0.77
     chimpan
    -0.76
     monop
    -0.76
     fel
    -0.76
    zees
    -0.75
     chim
    -0.75
     tam
    -0.75
     dur
    -0.74
     relig
    -0.74
     lin
    -0.73
    POSITIVE LOGITS
    Streamer
    1.14
    Getty
    1.11
    EPA
    1.08
    UTERS
    1.04
    Reuters
    0.99
    AAAAAAAA
    0.98
    DEBUG
    0.95
    Output
    0.92
    PRES
    0.91
    SEC
    0.91
    Act Density 0.145%

    No Known Activations