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

    Neuronpedia

    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    This neuron detects question phrases asking whether a particular action or practice is acceptable (e.g. “is it ok to…,” “is it bad practice to…,” “is it reasonable to use…?”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     synchronize
    -0.07
    datasets
    -0.06
     inspection
    -0.06
    amine
    -0.06
    زینه
    -0.06
    تيار
    -0.06
     레벨
    -0.06
    ระบ
    -0.06
    -resistant
    -0.06
    erialize
    -0.05
    POSITIVE LOGITS
     "=
    0.09
     Möglich
    0.07
    午
    0.07
     Rag
    0.07
    οκ
    0.07
     No
    0.07
    -animation
    0.07
     PyQt
    0.07
     Francie
    0.07
     herkes
    0.06
    Act Density 0.025%

    No Known Activations