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

    Neuronpedia

    INDEX
    Explanations

    phrases related to criminal accusations or legal matters

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to crime and justice

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     redes
    -0.57
     assorted
    -0.57
    eday
    -0.54
    accompan
    -0.54
     culminating
    -0.52
     bolstered
    -0.52
    ãĥĩãĤ£
    -0.52
    everal
    -0.50
    ãĥ¼ãĥĨãĤ£
    -0.49
     sandwic
    -0.49
    POSITIVE LOGITS
     anymore
    1.88
     nor
    1.82
    nor
    1.29
     whatsoever
    1.13
     anything
    1.00
    yet
    0.98
     unless
    0.98
     slightest
    0.97
     anywhere
    0.93
     :(
    0.92
    Act Density 0.838%

    No Known Activations