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

    Neuronpedia

    INDEX
    Explanations

    names of news organizations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     blur
    -0.69
    ''.
    -0.65
     charge
    -0.62
     stride
    -0.61
     shortage
    -0.61
     stricken
    -0.61
    ãĤ©
    -0.61
     perjury
    -0.61
     WRITE
    -0.60
     Dar
    -0.60
    POSITIVE LOGITS
    Games
    1.09
    Gold
    1.07
    Chem
    1.07
    Radio
    1.07
    Network
    1.05
    Down
    1.04
    swers
    1.04
    Detroit
    1.03
    Shop
    1.03
    Monitor
    1.01
    Act Density 0.163%

    No Known Activations