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

    Neuronpedia

    INDEX
    Explanations

    names of specific ancient naval vessels

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     fructose
    -1.04
     distressed
    -1.02
    ngth
    -1.00
     beet
    -1.00
     resemb
    -0.98
     ripe
    -0.94
    76561
    -0.93
     centrif
    -0.93
     stub
    -0.92
     tight
    -0.92
    POSITIVE LOGITS
    bush
    1.47
    itely
    1.37
    ois
    1.27
    fecture
    1.24
    stellar
    1.20
    idian
    1.19
    rences
    1.19
    ernal
    1.18
    ilver
    1.16
    quel
    1.15
    Act Density 1.160%

    No Known Activations