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

    Neuronpedia

    INDEX
    Explanations

    "'-"

    np_max-act · gemini-2.0-flash

    This neuron detects prime‐style position labels (e.g. “2′,” “4′”) in chemical compound names.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     dropout
    -0.07
     opioids
    -0.07
     SVM
    -0.06
    _IDS
    -0.06
    CLUS
    -0.06
    ARR
    -0.06
    �
    -0.06
     Fax
    -0.06
     BlockPos
    -0.06
    uby
    -0.06
    POSITIVE LOGITS
    리는
    0.07
    roat
    0.07
     altre
    0.07
    ران
    0.07
     string
    0.06
    ample
    0.06
    стро
    0.06
    чис
    0.06
    loon
    0.06
    handling
    0.06
    Act Density 0.002%

    No Known Activations