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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating completion or resolution

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate actions or decisions related to putting an end to something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Digest
    -0.67
     Provides
    -0.60
    etheless
    -0.57
     Dhabi
    -0.56
    Ł
    -0.56
    eming
    -0.55
     Recomm
    -0.55
     Associated
    -0.55
    =-=-=-=-
    -0.54
     Breakfast
    -0.54
    POSITIVE LOGITS
     aside
    0.90
     emphasis
    0.78
     jeopardy
    0.77
    channelAvailability
    0.77
     together
    0.76
     toget
    0.75
     brakes
    0.72
     lid
    0.71
    rescent
    0.69
     spotlight
    0.63
    Act Density 0.178%

    No Known Activations