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

    Neuronpedia

    INDEX
    Explanations

    no

    np_max-act · gemini-2.0-flash

    This neuron specifically detects the “assistant” speaker tag at the start of an assistant response.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    technical details related to processes and instructions.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    	buttons
    -0.07
     expire
    -0.06
     Numer
    -0.06
     Unc
    -0.06
    Uniform
    -0.06
     나타
    -0.06
     Succ
    -0.06
    Administrator
    -0.06
    Customers
    -0.06
    CHAIN
    -0.06
    POSITIVE LOGITS
    riteln
    0.06
    layıcı
    0.06
    (system
    0.06
     nem
    0.06
    oit
    0.06
     snel
    0.06
    (%
    0.06
    PELL
    0.06
     staple
    0.06
    fontName
    0.06
    Act Density 0.058%

    No Known Activations