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

    Neuronpedia

    INDEX
    Explanations

    quote

    np_max-act · gemini-2.0-flash

    Explanation of neuron 4 behavior: the main thing this neuron does is find instructions to quote or escape special characters (e.g. shell metacharacters)

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .toolStripButton
    -0.07
     pioneer
    -0.07
     cake
    -0.07
    خة
    -0.06
     بح
    -0.06
    丝
    -0.06
     UIScreen
    -0.06
    ури
    -0.06
    طبيق
    -0.06
     oval
    -0.06
    POSITIVE LOGITS
     şimdi
    0.06
     PY
    0.06
    Unable
    0.06
    .writeFileSync
    0.06
    	start
    0.06
     Thing
    0.06
     careful
    0.06
     unaware
    0.05
    GRA
    0.05
    Ve
    0.05
    Act Density 0.010%

    No Known Activations