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

    Neuronpedia

    INDEX
    Explanations

    the word "where" and its variations indicating location or context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    where followed by who/what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     simultaneously
    -0.43
     subtly
    -0.38
    Dış
    -0.38
     immediately
    -0.38
    archiviato
    -0.37
     completely
    -0.37
     Milit
    -0.37
     interesting
    -0.36
     simply
    -0.36
     easily
    -0.35
    POSITIVE LOGITS
     uſed
    0.54
     kasarigan
    0.48
    Wut
    0.47
     Cæsar
    0.47
     aDecoder
    0.47
     betweenstory
    0.46
     <>",
    0.46
     what
    0.46
     duquel
    0.45
     $_(
    0.45
    Act Density 0.014%

    No Known Activations