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

    Neuronpedia

    INDEX
    Explanations

    questions, especially those starting with the word "ask"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    Ĥ¬
    -0.73
    âĶĢâĶĢ
    -0.67
    cutting
    -0.66
    lim
    -0.65
     Scouting
    -0.65
    ccording
    -0.64
    edition
    -0.62
    swing
    -0.62
    absor
    -0.61
    zinski
    -0.61
    POSITIVE LOGITS
     rhet
    1.25
     questions
    1.21
     probing
    1.03
     forgiveness
    0.97
    naires
    0.97
     politely
    0.94
     plaint
    0.93
     Questions
    0.92
     permission
    0.91
     question
    0.84
    Act Density 0.415%

    No Known Activations