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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "say" often paired with different contexts or subjects

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express uncertainty or inability to make definitive statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    abwe
    -0.67
    folios
    -0.64
    figure
    -0.62
    onut
    -0.61
    velength
    -0.61
    poons
    -0.61
    ACTION
    -0.60
    llers
    -0.59
     livest
    -0.59
    pes
    -0.58
    POSITIVE LOGITS
     goodbye
    1.40
     Goodbye
    1.03
     definitively
    0.98
     aloud
    0.95
     hello
    0.93
     sorry
    0.87
     farewell
    0.84
     hi
    0.81
     bye
    0.80
     unequivocally
    0.78
    Act Density 0.067%

    No Known Activations