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

    Neuronpedia

    INDEX
    Explanations

    comparative adjectives and phrases indicating cost or effectiveness in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inho
    -0.14
    erer
    -0.14
    udu
    -0.14
    osh
    -0.14
    à¹Ħว
    -0.13
    iske
    -0.13
    ournal
    -0.13
    rossover
    -0.13
    erde
    -0.13
    æĹ
    -0.13
    POSITIVE LOGITS
     compared
    1.13
     Compared
    0.83
     compare
    0.58
    æ¯Ķ
    0.56
     comparison
    0.55
     relative
    0.52
     compares
    0.52
    compare
    0.47
     Compare
    0.47
     comparing
    0.46
    Act Density 0.335%

    No Known Activations