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

    Neuronpedia

    INDEX
    Explanations

    O

    np_max-act · gemini-2.0-flash

    This neuron detects the special header‐start token (“<|start_header_id|>”) that marks metadata sections in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    触
    -0.06
    ise
    -0.06
     �
    -0.06
     nhân
    -0.06
     wp
    -0.06
    _TYP
    -0.06
    Snap
    -0.06
     Terminator
    -0.06
     producer
    -0.06
    ationToken
    -0.06
    POSITIVE LOGITS
    .game
    0.07
    PO
    0.06
     gehen
    0.06
    (todo
    0.06
     pís
    0.06
     playoff
    0.06
     rápido
    0.06
     DirectoryInfo
    0.06
     soaring
    0.06
    -educated
    0.06
    Act Density 0.307%

    No Known Activations