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

    Neuronpedia

    INDEX
    Explanations

    words related to character names and interactions in a narrative context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    DOWN
    -0.77
    hement
    -0.73
    edin
    -0.73
    SPONSORED
    -0.71
    INAL
    -0.70
     largeDownload
    -0.70
    reement
    -0.69
    andowski
    -0.68
    Down
    -0.68
    owship
    -0.68
    POSITIVE LOGITS
    laws
    0.88
     unnoticed
    0.77
     virtue
    0.76
     proxy
    0.75
    products
    0.74
    product
    0.74
     stealth
    0.70
     Proxy
    0.68
    gone
    0.67
     leaps
    0.67
    Act Density 9.181%

    No Known Activations