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

    Neuronpedia

    INDEX
    Explanations

    phrases containing the word "Yes" and then, often, a statement or affirmation

    oai_token-act-pair · gpt-3.5-turbo

    repetitive affirmations or confirmations in dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    èĢħ
    -0.83
    arted
    -0.82
    Reviewed
    -0.76
    ļéĨĴ
    -0.73
    igion
    -0.72
    -+
    -0.70
    ende
    -0.69
    )=(
    -0.68
    è£ħ
    -0.68
    rance
    -0.68
    POSITIVE LOGITS
     sir
    1.04
     yes
    0.99
     yeah
    0.91
     technically
    0.87
     please
    0.81
     indeed
    0.81
     there
    0.81
     THERE
    0.79
     steroids
    0.76
     it
    0.76
    Act Density 0.076%

    No Known Activations