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

    Neuronpedia

    INDEX
    Explanations

    questions within the text

    oai_token-act-pair · gpt-3.5-turbo

    questions in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ikuman
    -0.98
     carbohyd
    -0.77
     transition
    -0.76
     referen
    -0.74
     corrid
    -0.73
     subur
    -0.73
     satell
    -0.72
     derog
    -0.70
     gobl
    -0.69
     cumbers
    -0.68
    POSITIVE LOGITS
     Answer
    1.55
     Does
    1.45
     Would
    1.44
     Should
    1.39
     Could
    1.36
     Wouldn
    1.34
    Answer
    1.34
     Surely
    1.31
     Probably
    1.27
     Are
    1.26
    Act Density 0.142%

    No Known Activations