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

    Neuronpedia

    INDEX
    Explanations

    #

    np_max-act · gemini-2.0-flash

    This neuron activates on lines that are configuration comments or section‐header separators (i.e. comment markers like “#” and adjacent punctuation) rather than actual setting names or values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    階
    -0.07
    рех
    -0.06
    	ts
    -0.06
    job
    -0.06
     bolest
    -0.06
    Customers
    -0.06
    -copy
    -0.06
    .CREATED
    -0.06
     canh
    -0.06
     bilder
    -0.06
    POSITIVE LOGITS
     phil
    0.07
     diminish
    0.06
     regarding
    0.06
     unsus
    0.06
     pcl
    0.06
    เ�
    0.06
     bringen
    0.06
     Swamp
    0.06
     midi
    0.06
     coincidence
    0.06
    Act Density 0.010%

    No Known Activations