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

    Neuronpedia

    INDEX
    Explanations

    terms indicating expertise or skill level

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

    savvy, smart, conscious, intelligent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    帖最后由
    -0.43
     parts
    -0.38
     Obat
    -0.38
    exitRule
    -0.35
    KommentareTeilen
    -0.34
    postId
    -0.34
     Buildings
    -0.33
     Pictures
    -0.32
    πομπ
    -0.32
     Parts
    -0.32
    POSITIVE LOGITS
     savvy
    1.89
    savvy
    1.66
     shrewd
    1.26
     astute
    1.13
     smart
    0.96
    smart
    0.93
     wise
    0.90
     clever
    0.89
    clever
    0.86
    Smart
    0.86
    Act Density 0.006%

    No Known Activations