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

    Neuronpedia

    INDEX
    Explanations

    numbers with a decimal point

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks, particularly periods

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     erection
    -0.70
     mixer
    -0.67
     bulky
    -0.67
     stocking
    -0.65
     performer
    -0.65
     bull
    -0.65
     trouble
    -0.65
     preventive
    -0.65
     pudding
    -0.64
     recorder
    -0.64
    POSITIVE LOGITS
    00
    1.23
    06
    1.12
    95
    1.10
    04
    1.10
    07
    1.09
    05
    1.08
    08
    1.07
    62
    1.06
    5
    1.06
    09
    1.05
    Act Density 0.081%

    No Known Activations