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

    Neuronpedia

    INDEX
    Explanations

    phrases related to intense or impactful actions or events

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to intensity or significance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Moroc
    -0.73
    WARN
    -0.65
    ã
    -0.65
     ÂŃ
    -0.60
    ijah
    -0.59
     Annex
    -0.58
    ãĥ¼ãĥĨ
    -0.57
    elfth
    -0.56
     Fas
    -0.56
     Instr
    -0.55
    POSITIVE LOGITS
     ;)
    1.13
     :)
    1.00
    !!!!
    0.99
     tho
    0.99
     alot
    0.98
     doesnt
    0.97
    !!!
    0.96
     :-)
    0.93
     !!
    0.92
     haha
    0.91
    Act Density 1.037%

    No Known Activations