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

    Neuronpedia

    INDEX
    Explanations

    prepositions or directional words related to physical movement

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ancial
    -0.83
     incent
    -0.79
    orld
    -0.77
    wcs
    -0.72
    discrimination
    -0.72
    GV
    -0.71
    oples
    -0.70
    iners
    -0.68
    interstitial
    -0.68
    Thousands
    -0.67
    POSITIVE LOGITS
     him
    1.02
     hers
    1.01
     Pyrrha
    0.98
     Jaune
    0.97
     her
    0.97
     Korra
    0.96
     Elsa
    0.91
     his
    0.90
     me
    0.88
     Chloe
    0.87
    Act Density 0.162%

    No Known Activations