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

    Neuronpedia

    INDEX
    Explanations

    phrases expressing extreme limits or overreaches in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ockey
    -0.15
    ussen
    -0.15
    pedia
    -0.15
    ãĤ¹ãĤ¿ãĥ¼
    -0.15
    mland
    -0.14
    æĽ´å¤ļ
    -0.14
    iffin
    -0.14
     defaultProps
    -0.14
    lus
    -0.14
    ienne
    -0.14
    POSITIVE LOGITS
     far
    0.34
     too
    0.28
     Too
    0.24
     Far
    0.24
    Too
    0.23
    Far
    0.23
     FAR
    0.21
    far
    0.21
     extreme
    0.21
    too
    0.20
    Act Density 0.019%

    No Known Activations