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

    Neuronpedia

    INDEX
    Explanations

    text embedded in special characters, '|<endoftext>|', and numbers

    oai_token-act-pair · gpt-3.5-turbo

    references to programming concepts or commands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Mechdragon
    -0.75
     Zig
    -0.70
     Ripple
    -0.64
     Scion
    -0.64
     Kinnikuman
    -0.62
     Cycle
    -0.61
     Ley
    -0.61
     Corpus
    -0.61
     Katy
    -0.61
     Xi
    -0.60
    POSITIVE LOGITS
    ³³³³³³³³³³³³³³³³
    0.81
    ³³³³³³³³
    0.80
    equal
    0.78
    ever
    0.77
    sonian
    0.77
    qual
    0.77
    fee
    0.74
    different
    0.73
    sav
    0.71
    herent
    0.71
    Act Density 0.078%

    No Known Activations