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

    Neuronpedia

    INDEX
    Explanations

    dates, particularly in the format "Month day" followed by the year

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "October."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    fman
    -0.73
    king
    -0.73
    cci
    -0.72
     Klux
    -0.68
    jri
    -0.68
    ever
    -0.67
    vet
    -0.67
     stewards
    -0.67
     attendant
    -0.66
    Magikarp
    -0.66
    POSITIVE LOGITS
     Surprise
    1.07
    å¹
    0.84
    flower
    0.81
     2018
    0.79
    nard
    0.78
    avia
    0.77
     2017
    0.76
     2014
    0.75
     2015
    0.73
    opus
    0.73
    Act Density 0.015%

    No Known Activations