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

    Neuronpedia

    INDEX
    Explanations

    phrases related to connection or relevance between subjects

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate relationships or connections to specific topics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Split
    -0.64
    oba
    -0.63
     Rush
    -0.62
    aez
    -0.61
     rush
    -0.60
     Splash
    -0.59
    osh
    -0.58
     Mist
    -0.58
     Sin
    -0.57
    eez
    -0.56
    POSITIVE LOGITS
     relating
    3.25
     pertaining
    2.29
     relate
    1.78
     relates
    1.68
     related
    1.68
     involving
    1.52
     arising
    1.51
     regarding
    1.45
     concerning
    1.40
     affecting
    1.40
    Act Density 0.012%

    No Known Activations