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

    Neuronpedia

    INDEX
    Explanations

    words related to conspiracy theories and controversial topics

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     constellation
    -1.01
     Hawaiian
    -0.96
     acquisitions
    -0.94
     LCS
    -0.93
     Iceland
    -0.93
    llan
    -0.90
     rings
    -0.89
    ADS
    -0.89
     Nigerian
    -0.87
     Roma
    -0.87
    POSITIVE LOGITS
    iest
    1.55
    ier
    1.53
    iers
    1.49
    estate
    1.45
    engers
    1.42
    uers
    1.40
    iness
    1.39
    erness
    1.38
    arella
    1.36
    orship
    1.35
    Act Density 1.356%

    No Known Activations