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

    Neuronpedia

    INDEX
    Explanations

    ASCII art and code

    np_max-act · gemini-2.0-flash

    This neuron detects special protocol or formatting tokens and structures—such as header delimiters, end‐of‐text markers, and code‐fence indicators—used to mark message boundaries and roles.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    IZER
    -0.08
    indrome
    -0.07
    _PIPELINE
    -0.06
    ves
    -0.06
    "To
    -0.06
     prototype
    -0.06
     kidnapped
    -0.06
    �
    -0.06
     Evropy
    -0.06
     Rodrigo
    -0.06
    POSITIVE LOGITS
    (hdc
    0.06
    nThe
    0.06
    入り
    0.06
    ItemSelectedListener
    0.06
     لو
    0.06
    ्मक
    0.06
     chỗ
    0.06
    addComponent
    0.06
    .Begin
    0.06
     전국
    0.06
    Act Density 0.020%

    No Known Activations