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

    Neuronpedia

    INDEX
    Explanations

    quantitative expressions related to amounts and values

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    æk
    -0.16
    loor
    -0.15
    ären
    -0.14
    RYPT
    -0.14
    ngine
    -0.14
    éļĨ
    -0.14
    Å¡t
    -0.14
     Troll
    -0.14
    orners
    -0.14
    omore
    -0.14
    POSITIVE LOGITS
     thousand
    0.59
     thousands
    0.56
     thou
    0.47
     thous
    0.47
     Thousand
    0.45
    -th
    0.45
     Thousands
    0.45
     TH
    0.40
    Thousands
    0.39
    åįĥ
    0.38
    Act Density 0.100%

    No Known Activations