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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "what" and related phrases indicating inquiry or emphasis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Anything
    -0.16
    éĤ£ç§į
    -0.16
    illo
    -0.15
    ë©
    -0.15
     anything
    -0.15
    á»ģn
    -0.15
    GIN
    -0.15
    ÙĤد
    -0.15
    anything
    -0.15
    klä
    -0.14
    POSITIVE LOGITS
     happens
    0.18
     we
    0.17
     happened
    0.16
    729
    0.15
     happen
    0.15
     happening
    0.15
     ultimately
    0.14
    ppe
    0.14
    zier
    0.14
    ãĥ³ãĥĸ
    0.14
    Act Density 0.027%

    No Known Activations