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

    Neuronpedia

    INDEX
    Explanations

    shape

    np_max-act · gemini-2.0-flash

    This neuron never activates—it doesn’t detect any particular token or pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .attrib
    -0.07
     asia
    -0.07
    =logging
    -0.07
    ظٹط
    -0.06
    	vertices
    -0.06
    FromBody
    -0.06
    BeginInit
    -0.06
    (alpha
    -0.06
     math
    -0.06
    @Inject
    -0.06
    POSITIVE LOGITS
    处
    0.08
     дви
    0.07
     Kathy
    0.07
     frameworks
    0.06
     TableView
    0.06
     Duncan
    0.06
     sentient
    0.06
     Compliance
    0.06
     sneak
    0.06
     Absolute
    0.06
    Act Density 0.000%

    No Known Activations