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

    Neuronpedia

    INDEX
    Explanations

    semicolons and parentheses

    np_max-act · gemini-2.0-flash

    This neuron detects C++ function declarations marked with “= delete,” i.e. deleted methods.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    -prop
    -0.07
    ₃
    -0.07
     accustomed
    -0.07
     Rational
    -0.07
     courageous
    -0.06
    -sup
    -0.06
     ideal
    -0.06
    umno
    -0.06
    -collapse
    -0.06
    ۲۰۰
    -0.06
    POSITIVE LOGITS
    addItem
    0.07
     skb
    0.07
     verdade
    0.06
     Whisper
    0.06
    aleb
    0.06
     strm
    0.06
     ImGui
    0.06
     mouseX
    0.06
     แก
    0.06
     yerde
    0.06
    Act Density 0.002%

    No Known Activations