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

    Neuronpedia

    INDEX
    Explanations

    references to time, events or comparisons related to "at least"

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "or at least" in sentences

    oai_token-act-pair · gpt-3.5-turboTriggered by @jordan
    New Auto-Interp
    Negative Logits
    iday
    -1.18
    regor
    -1.09
    hou
    -1.06
    enium
    -1.03
    iband
    -1.01
    pressed
    -0.98
    mage
    -0.98
    iple
    -0.98
    llular
    -0.97
    bris
    -0.97
    POSITIVE LOGITS
     insofar
    1.01
     partly
    0.97
     thats
    0.94
     unsurprisingly
    0.88
    etheless
    0.88
     Louie
    0.87
    orah
    0.87
     Leo
    0.86
     preferably
    0.85
     Thou
    0.85
    Act Density 0.321%

    No Known Activations