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

    Neuronpedia

    INDEX
    Explanations

    names of individuals

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     apologies
    -0.68
     representation
    -0.64
     equivalents
    -0.58
     compromise
    -0.57
     ambassadors
    -0.57
     conventions
    -0.56
     elimination
    -0.56
     Wonderland
    -0.55
     pleasure
    -0.55
     audiences
    -0.55
    POSITIVE LOGITS
    oglu
    1.13
    onen
    1.07
    chuk
    1.05
    hani
    1.02
    kov
    1.02
    zinski
    1.01
    gui
    1.01
    arov
    1.00
    iev
    0.99
    iani
    0.97
    Act Density 0.249%

    No Known Activations