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

    Neuronpedia

    INDEX
    Explanations

    proper nouns of people from diverse cultural backgrounds

    oai_token-act-pair · gpt-3.5-turbo

    names of notable figures and organizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     stitching
    -0.63
     redundancy
    -0.62
     recursive
    -0.61
     apologies
    -0.61
     magnification
    -0.60
     overload
    -0.60
     fixation
    -0.59
     constants
    -0.59
     multic
    -0.59
     confusing
    -0.58
    POSITIVE LOGITS
    ño
    0.94
    vati
    0.93
    ensis
    0.93
    gui
    0.90
    ouf
    0.88
    ahu
    0.88
    ÄŁ
    0.87
    pta
    0.87
    iev
    0.86
    angan
    0.86
    Act Density 0.386%

    No Known Activations