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

    Neuronpedia

    INDEX
    Explanations

    references to sandwich-related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to sandwiches

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tics
    -0.79
    mberg
    -0.75
    ibr
    -0.71
    aft
    -0.69
    orne
    -0.69
    negative
    -0.66
    effic
    -0.66
    pmwiki
    -0.66
    axis
    -0.65
    umen
    -0.63
    POSITIVE LOGITS
     sandwiches
    1.15
     sandwich
    1.04
    wich
    1.01
    bowl
    0.91
     salads
    0.91
     Sandwich
    0.87
     tray
    0.80
    anut
    0.80
     slices
    0.80
     salad
    0.79
    Act Density 0.012%

    No Known Activations