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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing strong emotions or confrontational statements

    oai_token-act-pair · gpt-3.5-turbo

    expressions of intention or decision-making

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    surprisingly
    -0.74
    etheless
    -0.72
     Cosponsors
    -0.70
    ItemTracker
    -0.64
    xtap
    -0.62
    ¿½
    -0.61
    ortium
    -0.61
     unsurprisingly
    -0.60
    ensibly
    -0.59
    Accessory
    -0.57
    POSITIVE LOGITS
     âĢ¦"
    1.41
    ,'"
    1.35
    ',"
    1.28
     ..."
    1.27
    âĢ¦"
    1.24
    '"
    1.24
    !'"
    1.24
    .")
    1.23
    ..."
    1.22
    .'"
    1.20
    Act Density 1.159%

    No Known Activations