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

    Neuronpedia

    INDEX
    Explanations

    references to specific countries

    oai_token-act-pair · gpt-3.5-turbo

    references to countries and companies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ories
    -0.74
    ests
    -0.66
    inations
    -0.66
     Kid
    -0.66
    ernels
    -0.64
     bolts
    -0.64
     Machines
    -0.63
     subtitles
    -0.63
     batteries
    -0.62
    inez
    -0.62
    POSITIVE LOGITS
     Called
    0.93
     whose
    0.78
    wide
    0.77
     LIKE
    0.77
    onym
    0.75
     resembling
    0.72
     Named
    0.72
     specializing
    0.71
     starved
    0.71
    atical
    0.69
    Act Density 0.332%

    No Known Activations