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

    Neuronpedia

    INDEX
    Explanations

    surprising or troubling situations

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ioxide
    -0.66
    gres
    -0.63
    ournal
    -0.62
    ahs
    -0.61
     bonded
    -0.61
     pesky
    -0.60
    urers
    -0.60
     govern
    -0.59
     governs
    -0.58
    ynthesis
    -0.58
    POSITIVE LOGITS
     considering
    1.06
     nonetheless
    1.04
     enough
    1.01
    ly
    0.99
     insofar
    0.91
     because
    0.88
     indeed
    0.87
    ingly
    0.83
     given
    0.82
     nevertheless
    0.78
    Act Density 0.141%

    No Known Activations