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

    Neuronpedia

    INDEX
    Explanations

    references to types of beer, specifically stout beers

    oai_token-act-pair · gpt-3.5-turbo

    references to specific types of beer, particularly stouts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     transmission
    -0.82
    chem
    -0.79
     Electronics
    -0.77
     transmit
    -0.77
    lev
    -0.76
     transmitting
    -0.75
     TRANS
    -0.74
    atically
    -0.70
    heat
    -0.70
    cells
    -0.69
    POSITIVE LOGITS
     Stout
    3.42
     stout
    2.23
     Fitzgerald
    1.24
     Tyr
    1.22
     Bold
    1.18
     Keen
    1.15
    ovan
    1.13
     Loch
    1.08
    WT
    1.08
     Brut
    1.01
    Act Density 0.067%

    No Known Activations