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

    Neuronpedia

    INDEX
    Explanations

    numeric values such as dates and measurements

    oai_token-act-pair · gpt-3.5-turbo

    instances of numerical values or quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     covenant
    -0.66
     nightly
    -0.64
     agre
    -0.61
     sculpt
    -0.59
     star
    -0.59
     camp
    -0.56
     lifes
    -0.56
     deaf
    -0.55
     cra
    -0.55
     dece
    -0.55
    POSITIVE LOGITS
    Advertisements
    1.20
    Conclusion
    1.02
    Advertisement
    1.01
    If
    1.00
    Instead
    0.98
    Whether
    0.97
    Translation
    0.96
    Official
    0.96
    Related
    0.94
    Besides
    0.94
    Act Density 0.735%

    No Known Activations