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

    Neuronpedia

    INDEX
    Explanations

    phrases related to putting an end to something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that suggest taking action or making changes in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fm
    -0.61
    arious
    -0.59
    comments
    -0.58
    ģ«
    -0.58
    arb
    -0.58
    Container
    -0.57
    planes
    -0.57
     unfocusedRange
    -0.57
    DI
    -0.57
     Admission
    -0.57
    POSITIVE LOGITS
     lid
    0.80
     aside
    0.78
     brakes
    0.78
    channelAvailability
    0.76
     toget
    0.74
     emphasis
    0.74
     together
    0.67
     disadvantage
    0.67
    ressed
    0.66
     jeopardy
    0.66
    Act Density 0.149%

    No Known Activations