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

    Neuronpedia

    INDEX
    Explanations

    words related to reversal or backwards movement

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "reverse."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Interstitial
    -0.94
    lished
    -0.77
    uay
    -0.72
    riers
    -0.71
    akov
    -0.71
    yers
    -0.67
     Trials
    -0.67
    %"
    -0.67
    liam
    -0.66
    utical
    -0.66
    POSITIVE LOGITS
     reverse
    1.11
     reversed
    0.98
     reversing
    0.96
    reverse
    0.93
     reversal
    0.88
     revers
    0.83
     chronological
    0.82
     flip
    0.81
     engineer
    0.79
     Reverse
    0.76
    Act Density 0.008%

    No Known Activations