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

    Neuronpedia

    INDEX
    Explanations

    workers

    np_max-act · gemini-2.0-flash

    The neuron responds to terms related to construction work (e.g., words like “construction,” “worker,” “site,” and “works”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     MADE
    -0.07
     most
    -0.07
     MOST
    -0.07
    LG
    -0.06
     Super
    -0.06
     rich
    -0.06
    ded
    -0.06
     Van
    -0.06
     gases
    -0.06
     Ed
    -0.06
    POSITIVE LOGITS
     příležit
    0.07
    òng
    0.06
     viscosity
    0.06
     lingering
    0.06
     steady
    0.06
     Debugger
    0.06
    िज
    0.06
     právní
    0.06
     Delegate
    0.06
     Reception
    0.06
    Act Density 0.017%

    No Known Activations