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

    Neuronpedia

    INDEX
    Explanations

    the word "preferred."

    oai_token-act-pair · gpt-3.5-turbo

    the word "preferred" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Tour
    -0.75
    arta
    -0.72
    á
    -0.72
    del
    -0.71
    bane
    -0.70
    akening
    -0.70
    amaz
    -0.70
    inas
    -0.69
    alien
    -0.68
    Breaking
    -0.68
    POSITIVE LOGITS
     preferred
    1.22
     embodiments
    0.90
     preferring
    0.81
     Preferred
    0.79
     prefers
    0.78
     plurality
    0.76
     favoured
    0.76
     endings
    0.76
     embodiment
    0.75
     favored
    0.75
    Act Density 0.005%

    No Known Activations