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

    Neuronpedia

    INDEX
    Explanations

    occurrences of numerals and symbols typically associated with structured data or codes

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

    hexadecimal addresses and identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <bos>
    -0.37
     conséquence
    -0.35
     następu
    -0.35
    спубли
    -0.35
     čierna
    -0.34
     المعيارى
    -0.34
     millón
    -0.33
     rozwiąz
    -0.33
    Xna
    -0.33
     gainera
    -0.33
    POSITIVE LOGITS
     kasarigan
    0.65
     utafitiHapana
    0.63
    onOptions
    0.56
     transfieras
    0.55
    ftagPool
    0.55
    MemoryWarning
    0.53
     cherchés
    0.52
    tagHelperRunner
    0.52
     UserDao
    0.51
    MLLoader
    0.51
    Act Density 0.001%

    No Known Activations