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

    Neuronpedia

    INDEX
    Explanations

    words related to specific intents or reasons

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to various types of purposes or intents

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Patriarch
    -0.88
     Ryder
    -0.76
     Ribbon
    -0.72
     Tens
    -0.69
     Sisters
    -0.68
     Beet
    -0.67
     Keys
    -0.66
     Clay
    -0.65
     sisters
    -0.64
     Trees
    -0.64
    POSITIVE LOGITS
     purposes
    0.92
    chwitz
    0.89
    tesy
    0.87
    ajor
    0.84
    ptions
    0.83
    farious
    0.82
    tenance
    0.80
    rative
    0.79
    folios
    0.78
    uty
    0.77
    Act Density 0.012%

    No Known Activations