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

    Neuronpedia

    INDEX
    Explanations

    507

    np_max-act · gemini-2.0-flash

    This neuron detects course prerequisite specifications, i.e. the alphanumeric course codes and numbers listed after “Prerequisite.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Applications
    -0.08
     Ubuntu
    -0.07
     circuit
    -0.07
    مند
    -0.07
     framework
    -0.07
    <Task
    -0.06
    "In
    -0.06
    ště
    -0.06
     Am
    -0.06
     Magn
    -0.06
    POSITIVE LOGITS
     línea
    0.06
    raries
    0.06
     нескольких
    0.06
     Kuzey
    0.06
    ==$
    0.06
     заверш
    0.06
    ATRIX
    0.06
    �
    0.06
    ooth
    0.06
    ітет
    0.06
    Act Density 0.004%

    No Known Activations