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

    Neuronpedia

    INDEX
    Explanations

    run

    np_max-act · gemini-2.0-flash

    This neuron activates on occurrences of the word “run,” particularly when it appears as a command or action in code or shell-instruction contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    theless
    -0.08
    	delta
    -0.06
     Wide
    -0.06
     distorted
    -0.06
    edList
    -0.06
    iele
    -0.06
    rored
    -0.06
    EEE
    -0.06
    belt
    -0.06
     stare
    -0.06
    POSITIVE LOGITS
     run
    0.08
    Run
    0.08
     n
    0.08
    _run
    0.08
    -use
    0.07
    -N
    0.07
    UN
    0.07
     conduct
    0.07
     Jon
    0.07
    ‌ان
    0.07
    Act Density 0.009%

    No Known Activations