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

    Neuronpedia

    INDEX
    Explanations

    references to newspapers

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ayne
    -0.71
    paren
    -0.71
    reek
    -0.69
    urden
    -0.66
     Defin
    -0.66
    onite
    -0.64
    asse
    -0.63
    fee
    -0.62
    heed
    -0.62
    itance
    -0.61
    POSITIVE LOGITS
     columnist
    0.99
     articles
    0.94
     editors
    0.90
     editor
    0.89
    letters
    0.87
     publisher
    0.87
     archives
    0.87
     clipping
    0.83
    lisher
    0.82
     newspapers
    0.81
    Act Density 0.041%

    No Known Activations