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

    Neuronpedia

    INDEX
    Explanations

    proper nouns related to technology or historical figures

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     pregn
    -0.93
    PsyNetMessage
    -0.85
     trave
    -0.76
     cryst
    -0.75
     Parenthood
    -0.75
     citiz
    -0.74
    perse
    -0.73
    blance
    -0.73
     conflic
    -0.72
     exha
    -0.72
    POSITIVE LOGITS
    strap
    1.31
    tails
    1.06
    roach
    1.04
    stra
    1.03
    warm
    1.00
    tail
    1.00
    ers
    0.96
    heimer
    0.95
    mith
    0.91
    borg
    0.90
    Act Density 7.782%

    No Known Activations