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

    Neuronpedia

    INDEX
    Explanations

    phrases related to deactivating, disabling, or switching off something

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to deactivation or turning off something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    mere
    -0.74
    Gall
    -0.73
    ĸļ
    -0.72
    ysical
    -0.71
    ollen
    -0.70
    ospace
    -0.70
    MER
    -0.70
    istani
    -0.69
    urd
    -0.68
    latest
    -0.67
    POSITIVE LOGITS
    screen
    0.79
     autop
    0.74
    ment
    0.71
     inhib
    0.69
     throttle
    0.69
     standby
    0.68
     leash
    0.67
     drugs
    0.66
    loading
    0.65
     alarms
    0.64
    Act Density 0.015%

    No Known Activations