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

    Neuronpedia

    INDEX
    Explanations

    he

    np_max-act · gemini-2.0-flash

    This neuron detects the special “header ID” marker tokens (e.g. <|start_header_id|>) in the OpenAI chat-format text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    erap
    -0.06
     Стар
    -0.06
    mh
    -0.06
    χω
    -0.06
     Satoshi
    -0.06
    Strength
    -0.06
     Removed
    -0.05
    WithString
    -0.05
    �
    -0.05
    سات
    -0.05
    POSITIVE LOGITS
    icons
    0.07
    _mem
    0.06
     clustering
    0.06
    Figure
    0.06
     hunting
    0.06
     outras
    0.06
    acağını
    0.06
     incomplete
    0.06
    -env
    0.06
     reverence
    0.06
    Act Density 0.019%

    No Known Activations