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

    Neuronpedia

    INDEX
    Explanations

    instances where attention is being focused or drawn to a particular subject

    oai_token-act-pair · gpt-3.5-turbo

    references to attention and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Tale
    -0.74
     Rebell
    -0.65
    nia
    -0.63
     Yugoslavia
    -0.63
    mos
    -0.62
    tein
    -0.61
     Dani
    -0.61
    lins
    -0.60
    iche
    -0.60
     Vers
    -0.59
    POSITIVE LOGITS
     spans
    0.96
     span
    0.95
     attention
    0.85
    orial
    0.84
    grab
    0.82
     largeDownload
    0.81
     seeker
    0.81
    estinal
    0.80
     seekers
    0.80
     grabbing
    0.77
    Act Density 0.053%

    No Known Activations