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

    Neuronpedia

    INDEX
    Explanations

    the root word "abstain" and its various endings

    oai_token-act-pair · gpt-4-turbo

    words and phrases related to abstaining or refraining from action

    oai_token-act-pair · gpt-3.5-turbo

    references to abstention or abstinence in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Sisters
    -0.75
    Reviewer
    -0.72
     Stevens
    -0.72
     Fitz
    -0.70
     Tycoon
    -0.70
     Spur
    -0.69
     Divinity
    -0.69
     Pioneer
    -0.68
    esan
    -0.68
    MER
    -0.68
    POSITIVE LOGITS
     abst
    1.15
    iencies
    1.00
    ention
    0.90
    aye
    0.86
    arians
    0.85
     scrut
    0.83
     minded
    0.81
    ree
    0.80
    itures
    0.80
    ensible
    0.76
    Act Density 0.005%

    No Known Activations