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

    Neuronpedia

    INDEX
    Explanations

    sentences where someone is speaking or quoting

    oai_token-act-pair · gpt-3.5-turbo

    reported speech or quotes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ILCS
    -0.74
    mental
    -0.73
    peg
    -0.72
     Protective
    -0.68
    against
    -0.68
    fighter
    -0.67
    pee
    -0.67
    Justice
    -0.67
    azines
    -0.66
    ghazi
    -0.66
    POSITIVE LOGITS
     goodbye
    1.11
     bluntly
    0.89
     anecd
    0.84
     farewell
    0.84
     boldly
    0.84
     unequivocally
    0.81
     confidently
    0.81
    escription
    0.79
     aloud
    0.78
     recently
    0.77
    Act Density 0.276%

    No Known Activations