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

    Neuronpedia

    INDEX
    Explanations

    instances of words related to investigation or surveillance, such as "sniffed out," "sighted," or "looking for."

    oai_token-act-pair · gpt-3.5-turbo

    instances of detection or identification, particularly involving animals or law enforcement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ributes
    -0.71
     Lifetime
    -0.71
     honors
    -0.69
    rative
    -0.68
    phabet
    -0.68
     regrets
    -0.67
     honorary
    -0.67
    assisted
    -0.67
    voc
    -0.66
    galitarian
    -0.65
    POSITIVE LOGITS
     lurking
    1.13
     suspicious
    1.06
     whereabouts
    1.04
     danger
    0.94
     clues
    0.94
     dangers
    0.90
     nearby
    0.88
     impending
    0.88
     intruder
    0.87
    ById
    0.86
    Act Density 0.666%

    No Known Activations