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

    Neuronpedia

    INDEX
    Explanations

    displays

    np_max-act · gemini-2.0-flash

    The neuron fires on document‐formatting markers and metadata tokens (e.g. the various <|…|> tags and header/footer identifiers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Zuk
    -0.07
    icích
    -0.06
     nodded
    -0.06
    -alist
    -0.06
     　 　 　 　 　 　
    -0.06
    Wednesday
    -0.06
     Aristotle
    -0.06
    adě
    -0.06
     yield
    -0.06
     achter
    -0.06
    POSITIVE LOGITS
    .='
    0.06
     Δεν
    0.06
     Devlet
    0.06
    ρό
    0.06
     ◎
    0.06
    ={()=>
    0.06
    ่ก
    0.06
     ILogger
    0.06
    čas
    0.06
    inars
    0.06
    Act Density 0.062%

    No Known Activations