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

    Neuronpedia

    INDEX
    Explanations

    personal pronouns 'you' or 'yourself' asking for opinions or thoughts

    oai_token-act-pair · gpt-3.5-turbo

    questions directed at the audience seeking opinions or thoughts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abad
    -0.77
    viol
    -0.68
     Izan
    -0.67
    atural
    -0.66
     Travels
    -0.64
    Journal
    -0.63
    ocol
    -0.61
    Lago
    -0.61
     Hels
    -0.60
    enges
    -0.58
    POSITIVE LOGITS
     think
    1.08
     reckon
    1.07
     guys
    1.07
     propose
    0.88
     suppose
    0.86
     say
    0.79
     mean
    0.78
     thoughts
    0.78
     expect
    0.76
     decide
    0.75
    Act Density 0.057%

    No Known Activations