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

    Neuronpedia

    INDEX
    Explanations

    complex programming syntax and symbols

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    scientific references and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ValueStyle
    -0.98
     kasarigan
    -0.97
    OGND
    -0.95
    Демографія
    -0.92
    :✨
    -0.91
    <unused74>
    -0.90
     فريبيس
    -0.90
    <pad>
    -0.90
    <unused8>
    -0.89
    <unused14>
    -0.89
    POSITIVE LOGITS
     I
    0.36
    .
    0.36
    ↵
    0.36
     are
    0.34
    ib
    0.33
    ↵↵
    0.31
     have
    0.31
     etc
    0.29
     try
    0.28
     A
    0.28
    Act Density 0.596%

    No Known Activations