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

    Neuronpedia

    INDEX
    Explanations

    Mem

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the prefix “Mem” at the start of words.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Wald
    -0.07
     Aly
    -0.07
    itch
    -0.07
     Guy
    -0.07
     Arthur
    -0.07
     starch
    -0.07
     Albania
    -0.07
     Wild
    -0.06
     Stuart
    -0.06
    ,width
    -0.06
    POSITIVE LOGITS
     Memo
    0.10
     memo
    0.09
    Memo
    0.09
     Mem
    0.09
     mem
    0.09
     Memor
    0.08
    Mem
    0.07
     ConcurrentHashMap
    0.07
     MEM
    0.07
     memorandum
    0.07
    Act Density 0.011%

    No Known Activations