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

    Neuronpedia

    INDEX
    Explanations

    names, such as names of places, people, or products

    oai_token-act-pair · gpt-3.5-turbo

    common suffixes or endings of words

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     mete
    -0.61
    accompan
    -0.60
     Paran
    -0.59
     Carib
    -0.58
     Investig
    -0.55
     Rats
    -0.55
     cannabin
    -0.55
     Tonight
    -0.54
     Nare
    -0.54
    Trend
    -0.53
    POSITIVE LOGITS
     Pradesh
    0.86
    orage
    0.83
    iencies
    0.83
    urnal
    0.79
    ilk
    0.77
    wered
    0.76
    sembly
    0.76
    phrine
    0.75
    hed
    0.74
    nikov
    0.72
    Act Density 0.220%

    No Known Activations