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

    Neuronpedia

    INDEX
    Explanations

    recipes with detailed instructions for baking and cooking

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     affili
    -0.81
     treasury
    -0.76
     ancestor
    -0.75
     prohib
    -0.73
     dating
    -0.73
     evidenced
    -0.71
     motiv
    -0.71
     affiliated
    -0.71
     estranged
    -0.70
     ticket
    -0.70
    POSITIVE LOGITS
     Then
    1.27
    Then
    1.26
    Remove
    1.26
     Slowly
    1.16
    Once
    1.07
    Repeat
    1.05
    then
    1.04
     Afterwards
    1.04
    Optional
    1.01
     Fold
    1.01
    Act Density 0.132%

    No Known Activations