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

    Neuronpedia

    INDEX
    Explanations

    written content

    np_max-act · gemini-2.0-flash

    This neuron never activates on any regular text—it’s essentially a dead neuron that doesn’t detect any tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Tracking
    -0.07
    haf
    -0.06
     gül
    -0.06
     heartbeat
    -0.06
    (period
    -0.06
     prog
    -0.06
    adam
    -0.06
    uber
    -0.06
     notions
    -0.06
     Toe
    -0.06
    POSITIVE LOGITS
     "*
    0.07
    ={`/
    0.07
     isKindOfClass
    0.07
     вы
    0.07
    _needed
    0.07
     elektrik
    0.06
    ="{
    0.06
    ्रक
    0.06
    .Serialization
    0.06
     đăng
    0.06
    Act Density 0.009%

    No Known Activations