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

    Neuronpedia

    INDEX
    Explanations

    chronological context

    np_max-act · gemini-2.0-flash

    The neuron detects numeric temporal expressions—number + time‐unit phrases like “Three days,” “months after,” or “years later.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     org
    -0.07
     Since
    -0.07
    .connect
    -0.06
    -file
    -0.06
     pys
    -0.06
     sauce
    -0.06
     mechanisms
    -0.06
     MessageType
    -0.06
     sorts
    -0.06
     holes
    -0.06
    POSITIVE LOGITS
    igham
    0.06
     watt
    0.06
     cấu
    0.06
    ウィ
    0.06
     "");
↵
    0.06
    なんて
    0.06
     Irvine
    0.06
     "[
    0.06
    しても
    0.06
    -consuming
    0.06
    Act Density 0.058%

    No Known Activations