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

    Neuronpedia

    INDEX
    Explanations

    phrases describing something as ideal or suitable for specific situations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating suitability or perfection for various scenarios

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chenko
    -0.71
    wine
    -0.68
    book
    -0.68
    allah
    -0.65
    owned
    -0.64
    oller
    -0.64
     tallied
    -0.64
    bil
    -0.63
     Accessed
    -0.63
    $,
    -0.63
    POSITIVE LOGITS
    geries
    1.04
     example
    0.96
    bidden
    0.90
     purposes
    0.86
    gotten
    0.84
    gery
    0.83
    WARD
    0.80
     beginners
    0.78
    gettable
    0.77
     awhile
    0.76
    Act Density 0.139%

    No Known Activations