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

    Neuronpedia

    INDEX
    Explanations

    phrases or sentences describing the composition or makeup of things

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate composition or the elements that make up something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hawks
    -0.64
    abad
    -0.63
    hello
    -0.62
    Leaks
    -0.61
    talk
    -0.60
    Lear
    -0.59
    ESE
    -0.59
    mort
    -0.56
    EMS
    -0.56
     messenger
    -0.55
    POSITIVE LOGITS
     mainly
    1.15
     chiefly
    1.11
     primarily
    1.11
     solely
    1.06
     entirely
    1.05
     principally
    1.05
     mostly
    1.02
     predominantly
    0.99
     largely
    0.96
     of
    0.94
    Act Density 0.056%

    No Known Activations