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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    This neuron detects numeric literals (including integers and floats) in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     resurgence
    -0.06
     Reload
    -0.06
    орош
    -0.06
    当然
    -0.06
    _estimators
    -0.06
    PropertyChanged
    -0.06
    вая
    -0.06
    อาจ
    -0.06
     crib
    -0.06
     всп
    -0.06
    POSITIVE LOGITS
    Bre
    0.07
     unions
    0.07
     wall
    0.06
    alion
    0.06
     ngay
    0.06
     milli
    0.06
    antal
    0.06
     insn
    0.06
     FORCE
    0.06
    μεν
    0.06
    Act Density 0.060%

    No Known Activations