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

    Neuronpedia

    INDEX
    Explanations

    phrases related to private or personal details

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    corn
    -1.02
    dule
    -0.84
    udeb
    -0.83
    soDeliveryDate
    -0.77
    ichick
    -0.76
    papers
    -0.75
    sein
    -0.72
    zeb
    -0.71
    enburg
    -0.70
    sup
    -0.69
    POSITIVE LOGITS
     intimate
    0.98
    ewitness
    0.82
     acquainted
    0.76
     details
    0.74
     experience
    0.73
     pleasure
    0.72
     familiarity
    0.72
     priv
    0.71
     IMAGES
    0.70
     sensations
    0.70
    Act Density 0.103%

    No Known Activations