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

    Neuronpedia

    INDEX
    Explanations

    instances of missed opportunities or potential outcomes

    oai_token-act-pair · gpt-3.5-turbo

    phrases expressing unrealized possibilities or hypothetical scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     muse
    -0.76
    mits
    -0.60
    isms
    -0.59
     forget
    -0.58
     believe
    -0.57
    ism
    -0.55
     recall
    -0.55
    quart
    -0.55
    reciation
    -0.54
     realize
    -0.54
    POSITIVE LOGITS
     been
    1.19
    been
    1.11
     Been
    1.05
     resulted
    1.04
     lasted
    0.96
     prevented
    0.96
     benefited
    0.95
     mattered
    0.92
     arisen
    0.89
     occurred
    0.89
    Act Density 0.069%

    No Known Activations