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

    Neuronpedia

    INDEX
    Explanations

    the phrase "phrase" occurring in various contexts and lengths

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases or expressions in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    elsen
    -0.81
     Flavoring
    -0.74
     Skydragon
    -0.74
    gewater
    -0.70
    DERR
    -0.69
    ÄŁ
    -0.69
    hari
    -0.66
    ntil
    -0.64
    gur
    -0.64
     Democr
    -0.64
    POSITIVE LOGITS
    phrase
    1.11
     phrase
    1.06
     phrases
    1.02
    ology
    0.99
    terday
    0.88
    naire
    0.87
     uttered
    0.87
    stress
    0.83
    witz
    0.81
     aloud
    0.81
    Act Density 0.012%

    No Known Activations