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

    Neuronpedia

    INDEX
    Explanations

    phrases related to distinguishable or notable characteristics

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Other."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    olt
    -0.78
    udence
    -0.74
    ttle
    -0.74
    opus
    -0.73
    iste
    -0.67
     anew
    -0.67
    ocado
    -0.66
    aternity
    -0.65
    '"
    -0.65
    ony
    -0.64
    POSITIVE LOGITS
    worldly
    1.58
    wise
    1.33
     than
    1.28
     factors
    1.11
     notable
    1.10
     considerations
    1.08
    quickShipAvailable
    1.07
     examples
    1.02
     possibilities
    1.01
     noteworthy
    0.99
    Act Density 0.053%

    No Known Activations