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

    Neuronpedia

    INDEX
    Explanations

    exclamation marks

    np_max-act · gemini-2.0-flash

    This neuron responds to logical negation (‘!’) operators in code.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     derivative
    -0.07
     RF
    -0.06
     seven
    -0.06
    μά
    -0.06
     TSA
    -0.06
     李
    -0.06
    agara
    -0.06
    -neutral
    -0.06
     ominous
    -0.06
    าอ
    -0.06
    POSITIVE LOGITS
    [word
    0.07
     Emb
    0.07
    collector
    0.06
     {})↵
    0.06
     Scout
    0.06
    #g
    0.06
    /csv
    0.06
    EATURE
    0.06
    "]);
↵
    0.06
    [${
    0.06
    Act Density 0.002%

    No Known Activations